Corpus

We develop conjectures and theorems expressing the idea that the prime sequence exhibits computational irreducibility in the transition from one prime to its successor. Informally, given a prime pp p, no general algorithm can compute the least prime greater than pp p substantially faster than sequentially testing candidates for primality, except possibly on sparse input sets. Our framework proceeds along complementary lines. First, we formalize Prime Successor Irreducibility in a Turing-machine complexity model (PSI-T), asserting lower bounds on running time relative to a sequential baseline. Second, we propose a Kolmogorov-complexity formulation (PSI-K), asserting that typical prime gaps are algorithmically incompressible at their scale; we prove PSI-K(c, $δ$) unconditionally for all fixed c<1 using standard sieve bounds. Third, we develop weakness-based formulations: PSI-W (sparse-set anti-concentration) shows no small menu of gap values captures a noticeable fraction of primes, while PSI-W-LE shows collision probabilities decay and logical entropy tends to 1. These extend to prime constellations and consecutive gap vectors. Finally, a sieve-theoretic framework connects local obstruction patterns to Selberg weakness parameters. The PSI-K and weakness formulations connect irreducibility to classical statistical questions about prime gaps. Using the relationship between Kolmogorov complexity and Shannon entropy, we derive rigorous lower bounds on prime gap entropy in dyadic intervals [X,2X]. Together, these formulations provide a unified complexity-theoretic perspective on the apparent local unpredictability of the prime sequence, without asserting randomness or independence.

Motivational architectures in cognitive AI have largely been designed for physical agents regulating bodily needs. Conversational agents operate in a different regime: their sensorimotor loop is linguistic, their environment is a user's evolving mental state, and their consequential actions are speech acts, tool invocations, and strategic silences. This paper proposes a conversational reinterpretation of the OpenPsi motivational lineage, coupled to MetaMo's higher-level motivational scaffold, for agents built on a modular execution substrate. Homeostasis is recast in dialogue-native terms: the agent regulates competence, uncertainty reduction, affiliation, affinity, legitimacy, nurturing, and aesthetic coherence rather than bodily deficits. We propose three contributions: a ten-stage motivational processing pipeline that architecturally separates cognitive modulation from situational appraisal; a dual decision strategy blending urgency-driven fast response with deliberative multi-goal optimization; and an architecturally useful distinction between pre-action feelings and post-action emotions as functionally different forms of affect. We specialize the framework to two example agents -- CompanionAgent and ResearchAgent -- and sketch its extension to social robotics and domain-generic human-level AGI.

We develop conjectures and theorems expressing the idea that the prime sequence exhibits computational irreducibility in the transition from one prime to its successor. Informally, given a prime pp p, no general algorithm can compute the least prime greater than pp p substantially faster than sequentially testing candidates for primality, except possibly on sparse input sets. Our framework proceeds along complementary lines. First, we formalize Prime Successor Irreducibility in a Turing-machine complexity model (PSI-T), asserting lower bounds on running time relative to a sequential baseline. Second, we propose a Kolmogorov-complexity formulation (PSI-K), asserting that typical prime gaps are algorithmically incompressible at their scale; we prove PSI-K(c, $δ$) unconditionally for all fixed c<1 using standard sieve bounds. Third, we develop weakness-based formulations: PSI-W (sparse-set anti-concentration) shows no small menu of gap values captures a noticeable fraction of primes, while PSI-W-LE shows collision probabilities decay and logical entropy tends to 1. These extend to prime constellations and consecutive gap vectors. Finally, a sieve-theoretic framework connects local obstruction patterns to Selberg weakness parameters. The PSI-K and weakness formulations connect irreducibility to classical statistical questions about prime gaps. Using the relationship between Kolmogorov complexity and Shannon entropy, we derive rigorous lower bounds on prime gap entropy in dyadic intervals [X,2X]. Together, these formulations provide a unified complexity-theoretic perspective on the apparent local unpredictability of the prime sequence, without asserting randomness or independence.

Motivational architectures in cognitive AI have largely been designed for physical agents regulating bodily needs. Conversational agents operate in a different regime: their sensorimotor loop is linguistic, their environment is a user's evolving mental state, and their consequential actions are speech acts, tool invocations, and strategic silences. This paper proposes a conversational reinterpretation of the OpenPsi motivational lineage, coupled to MetaMo's higher-level motivational scaffold, for agents built on a modular execution substrate. Homeostasis is recast in dialogue-native terms: the agent regulates competence, uncertainty reduction, affiliation, affinity, legitimacy, nurturing, and aesthetic coherence rather than bodily deficits. We propose three contributions: a ten-stage motivational processing pipeline that architecturally separates cognitive modulation from situational appraisal; a dual decision strategy blending urgency-driven fast response with deliberative multi-goal optimization; and an architecturally useful distinction between pre-action feelings and post-action emotions as functionally different forms of affect. We specialize the framework to two example agents -- CompanionAgent and ResearchAgent -- and sketch its extension to social robotics and domain-generic human-level AGI.

This paper introduces ActPC-Geom, an approach to accelerate Active Predictive Coding (ActPC) in neural networks by integrating information geometry, specifically using Wasserstein-metric-based methods for measure-dependent gradient flows. We propose replacing KL-divergence in ActPC's predictive error assessment with the Wasserstein metric, suggesting this may enhance network robustness. To make this computationally feasible, we present strategies including: (1) neural approximators for inverse measure-dependent Laplacians, (2) approximate kernel PCA embeddings for low-rank approximations feeding into these approximators, and (3) compositional hypervector embeddings derived from kPCA outputs, with algebra optimized for fuzzy FCA lattices learned through neural architectures analyzing network states. This results in an ActPC architecture capable of real-time online learning and integrating continuous (e.g., transformer-like or Hopfield-net-like) and discrete symbolic ActPC networks, including frameworks like OpenCog Hyperon or ActPC-Chem for algorithmic chemistry evolution. Shared probabilistic, concept-lattice, and hypervector models enable symbolic-subsymbolic integration. Key features include (1) compositional reasoning via hypervector embeddings in transformer-like architectures for tasks like commonsense reasoning, and (2) Hopfield-net dynamics enabling associative long-term memory and attractor-driven cognitive features. We outline how ActPC-Geom combines few-shot learning with online weight updates, enabling deliberative thinking and seamless symbolic-subsymbolic reasoning. Ideas from Galois connections are explored for efficient hybrid ActPC/ActPC-Chem processing. Finally, we propose a specialized HPC design optimized for real-time focused attention and deliberative reasoning tailored to ActPC-Geom's demands.

This paper addresses the problem of formalizing and quantifying the concept of "intensional inheritance" between two concepts. We begin by conceiving the intensional inheritance of $W$ from $F$ as the amount of information the proposition "x is $F$ " provides about the proposition "x is $W$. To flesh this out, we consider concepts $F$ and $W$ defined by sets of properties $\left\{F_{1}, F_{2}, \ldots, F_{n}\right\}$ and $\left\{W_{1}, W_{2}, \ldots, W_{m}\right\}$ with associated degrees $\left\{d_{1}, d_{2}, \ldots, d_{n}\right\}$ and $\left\{e_{1}, e_{2}, \ldots, e_{m}\right\}$, respectively, where the properties may overlap. We then derive formulas for the intensional inheritance using both Shannon information theory and algorithmic information theory, incorporating interaction information among properties. We examine a special case where all properties are mutually exclusive and calculate the intensional inheritance in this case in both frameworks. We also derive expressions for $P(W \mid F)$ based on the mutual information formula. Finally we consider the relationship between intensional inheritance and conventional set-theoretic "extensional" inheritance, concluding that in our information-theoretic framework, extensional inheritance emerges as a special case of intensional inheritance.

We propose a robust transformer architecture designed to prevent prompt injection attacks and ensure secure, reliable response generation. Our PICO (Prompt Isolation and Cybersecurity Oversight) framework structurally separates trusted system instructions from untrusted user inputs through dual channels that are processed independently and merged only by a controlled, gated fusion mechanism. In addition, we integrate a specialized Security Expert Agent within a Mixture-of-Experts (MoE) framework and incorporate a Cybersecurity Knowledge Graph (CKG) to supply domain-specific reasoning. Our training design further ensures that the system prompt branch remains immutable while the rest of the network learns to handle adversarial inputs safely. This PICO framework is presented via a general mathematical formulation, then elaborated in terms of the specifics of transformer architecture, and fleshed out via hypothetical case studies including Policy Puppetry attacks. While the most effective implementation may involve training transformers in a PICO-based way from scratch, we also present a cost-effective fine-tuning approach.

We present a lattice-based scheme for homomorphic evaluation of quantum programs and proofs that remains secure against quantum adversaries. Classical homomorphic encryption is lifted to the quantum setting by replacing composite-order groups with Module Learning-With-Errors (MLWE) lattices and by generalizing polynomial functors to bounded natural super functors (BNSFs). A secret depolarizing BNSF mask hides amplitudes, while each quantum state is stored as an MLWE ciphertext pair. We formalize security with the qIND-CPA game that allows coherent access to the encryption oracle and give a four-hybrid reduction to decisional MLWE. The design also covers practical issues usually left open. A typed QC-bridge keeps classical bits produced by measurements encrypted yet still usable as controls, with weak-measurement semantics for expectation-value workloads. Encrypted Pauli twirls add circuit privacy. If a fixed knowledge base is needed, its axioms are shipped as MLWE "capsules"; the evaluator can use them but cannot read them. A rho-calculus driver schedules encrypted tasks across several QPUs and records an auditable trace on an RChain-style ledger. Performance analysis shows that the extra lattice arithmetic fits inside today's QPU idle windows: a 100-qubit, depth-10^3 teleportation-based proof runs in about 10 ms, the public key (seed only) is 32 bytes, and even a CCA-level key stays below 300 kB. A photonic Dirac-3 prototype that executes homomorphic teleportation plus knowledge-base-relative amplitude checks appears feasible with current hardware. These results indicate that fully homomorphic, knowledge-base-aware quantum reasoning is compatible with near-term quantum clouds and standard post-quantum security assumptions.

We present a proof architecture for \(P \neq NP\) based on an upper--lower clash in polytime-capped conditional description length. We construct an efficiently samplable family of SAT instances \(Y\) such that every satisfying witness for \(Y\) yields the same global message \(M(Y)\). If \(P=NP\), then a standard polynomial-time SAT self-reduction recovers \(M(Y)\) from \(Y\), so \[ K_{\mathrm{poly}}(M(Y)\mid Y)=O(1). \] The lower-bound side shows the opposite. For the same ensemble, no fixed polynomial-time observer can gain substantial predictive advantage on a linear number of selected message coordinates. The argument treats computation as an evidence-producing process: predictive advantage is converted into constructible-dual evidence skew and then into pairwise distinctions between message-opposite worlds. A normalization theorem shows that every target-relevant non-neutral evidence leaf is either a safe-buffer observation or a hidden-gauge observation. Safe-buffer observations have negligible leakage, while hidden-gauge observations are limited by gauge-rank accounting. This yields an atomic evidence budget implying that total message-resolving advantage is \(o(t)\) across \(t\) selected coordinates. Boundary-law mixing gives the near-random baseline for the visible surface. Combining this with the evidence budget gives product small-success and then, by Compression-from-Success, \[ K_{\mathrm{poly}}(M(Y)\mid Y)\ge Ω(t) \] with high probability. This contradicts the constant upper bound from \(P=NP\). Therefore \(P \neq NP\).

We explore a novel paradigm (labeled ActPC-Chem) for biologically inspired, goal-guided artificial intelligence (AI) centered on a form of Discrete Active Predictive Coding (ActPC) operating within an algorithmic chemistry of rewrite rules. ActPC-Chem is envisioned as a foundational "cognitive kernel" for advanced cognitive architectures, such as the OpenCog Hyperon system, incorporating essential elements of the PRIMUS cognitive architecture. The central thesis is that general-intelligence-capable cognitive structures and dynamics can emerge in a system where both data and models are represented as evolving patterns of metagraph rewrite rules, and where prediction errors, intrinsic and extrinsic rewards, and semantic constraints guide the continual reorganization and refinement of these rules. Using a virtual "robot bug" thought experiment, we illustrate how such a system might self-organize to handle challenging tasks involving delayed and context-dependent rewards, integrating causal rule inference (AIRIS) and probabilistic logical abstraction (PLN) to discover and exploit conceptual patterns and causal constraints. Next, we describe how continuous predictive coding neural networks, which excel at handling noisy sensory data and motor control signals, can be coherently merged with the discrete ActPC substrate. Finally, we outline how these ideas might be extended to create a transformer-like architecture that foregoes traditional backpropagation in favor of rule-based transformations guided by ActPC. This layered architecture, supplemented with AIRIS and PLN, promises structured, multi-modal, and logically consistent next-token predictions and narrative sequences.

We articulate here a series of specific metagoals designed to address the challenge of creating AGI systems that possess the ability to flexibly self-modify yet also have the propensity to maintain key invariant properties of their goal systems 1) a series of goal-stability metagoals aimed to guide a system to a condition in which goal-stability is compatible with reasonably flexible self-modification 2) a series of moderated-goal-evolution metagoals aimed to guide a system to a condition in which control of the pace of goal evolution is compatible with reasonably flexible self-modification The formulation of the metagoals is founded on fixed-point theorems from functional analysis, e.g. the Contraction Mapping Theorem and constructive approximations to Schauder's Theorem, applied to probabilistic models of system behavior We present an argument that the balancing of self-modification with maintenance of goal invariants will often have other interesting cognitive side-effects such as a high degree of self understanding Finally we argue for the practical value of a hybrid metagoal combining moderated-goal-evolution with pursuit of goal-stability -- along with potentially other metagoals relating to goal-satisfaction, survival and ongoing development -- in a flexible fashion depending on the situation

We provide a comparative analysis of the deduction, induction, and abduction formulas used in Probabilistic Logic Networks (PLN) and the Non-Axiomatic Reasoning System (NARS), two uncertain reasoning frameworks aimed at AGI. One difference between the two systems is that, at the level of individual inference rules, PLN directly leverages both term and relationship probabilities, whereas NARS only leverages relationship frequencies and has no simple analogue of term probabilities. Thus we focus here on scenarios where there is high uncertainty about term probabilities, and explore how this uncertainty influences the comparative inferential conclusions of the two systems. We compare the product of strength and confidence ($s\times c$) in PLN against the product of frequency and confidence ($f\times c$) in NARS (quantities we refer to as measuring the "power" of an uncertain statement) in cases of high term probability uncertainty, using heuristic analyses and elementary numerical computations. We find that in many practical situations with high term probability uncertainty, PLN and NARS formulas give very similar results for the power of an inference conclusion, even though they sometimes come to these similar numbers in quite different ways.

We present an operational semantics for the language MeTTa.

A moderately detailed consideration of interactive LLMs as cognitive systems is given, focusing on LLMs circa mid-2023 such as ChatGPT, GPT-4, Bard, Llama, etc.. Cognitive strengths of these systems are reviewed, and then careful attention is paid to the substantial differences between the sort of cognitive system these LLMs are, and the sort of cognitive systems human beings are. It is found that many of the practical weaknesses of these AI systems can be tied specifically to lacks in the basic cognitive architectures according to which these systems are built. It is argued that incremental improvement of such LLMs is not a viable approach to working toward human-level AGI, in practical terms given realizable amounts of compute resources. This does not imply there is nothing to learn about human-level AGI from studying and experimenting with LLMs, nor that LLMs cannot form significant parts of human-level AGI architectures that also incorporate other ideas. Social and ethical matters regarding LLMs are very briefly touched from this perspective, which implies that while care should be taken regarding misinformation and other issues, and economic upheavals will need their own social remedies based on their unpredictable course as with any powerfully impactful technology, overall the sort of policy needed as regards modern LLMs is quite different than would be the case if a more credible approximation to human-level AGI were at hand.

This report summarizes a short study of the performance of GPT-4 on the ETHICS dataset. The ETHICS dataset consists of five sub-datasets covering different fields of ethics: Justice, Deontology, Virtue Ethics, Utilitarianism, and Commonsense Ethics. The moral judgments were curated so as to have a high degree of agreement with the aim of representing shared human values rather than moral dilemmas. GPT-4's performance is much better than that of previous models and suggests that learning to work with common human values is not the hard problem for AI ethics.

An introduction to the OpenCog Hyperon framework for Artificiai General Intelligence is presented. Hyperon is a new, mostly from-the-ground-up rewrite/redesign of the OpenCog AGI framework, based on similar conceptual and cognitive principles to the previous OpenCog version, but incorporating a variety of new ideas at the mathematical, software architecture and AI-algorithm level. This review lightly summarizes: 1) some of the history behind OpenCog and Hyperon, 2) the core structures and processes underlying Hyperon as a software system, 3) the integration of this software system with the SingularityNET ecosystem's decentralized infrastructure, 4) the cognitive model(s) being experimentally pursued within Hyperon on the hopeful path to advanced AGI, 5) the prospects seen for advanced aspects like reflective self-modification and self-improvement of the codebase, 6) the tentative development roadmap and various challenges expected to be faced, 7) the thinking of the Hyperon team regarding how to guide this sort of work in a beneficial direction ... and gives links and references for readers who wish to delve further into any of these aspects.

We introduce a formal meta-language for probabilistic programming, capable of expressing both programs and the type systems in which they are embedded. We are motivated here by the desire to allow an AGI to learn not only relevant knowledge (programs/proofs), but also appropriate ways of reasoning (logics/type systems). We draw on the frameworks of cubical type theory and dependent typed metagraphs to formalize our approach. In doing so, we show that specific constructions within the meta-language can be related via bisimulation (implying path equivalence) to the type systems they correspond. This allows our approach to provide a convenient means of deriving synthetic denotational semantics for various type systems. Particularly, we derive bisimulations for pure type systems (PTS), and probabilistic dependent type systems (PDTS). We discuss further the relationship of PTS to non-well-founded set theory, and demonstrate the feasibility of our approach with an implementation of a bisimulation proof in a Guarded Cubical Type Theory type checker.

Problem of reliable democratic governance is critical for survival of any community, and it will be critical for communities powered with Artificial Intelligence (AI) systems upon developments of the latter. Apparently, it will be getting more and more critical because of increasing speeds and scales of electronic communications and decreasing latencies in system responses. In order to address this need, we present design and implementation of a reputation system supporting "liquid democracy" principle. The system is based on "weighted liquid rank" algorithm employing different sorts of explicit and implicit ratings being exchanged by members of the society as well as implicit assessments of of the members based on measures of their activity in the society. The system is evaluated against live social network data with help of simulation modelling for an online marketplace case.

We simulate a reputation system in a market to optimise the balance between market security and market equity. We introduce a method of using a reputation system that will stabilise the distribution of wealth in a market in a fair manner. We also introduce metrics of a modified Gini that takes production quality into account, a way to use a weighted Pearson as a tool to optimise balance.

Abstract We present an agent based simulation model configured for exploring the dynamics of disease spread in the context of agents that group together through homophily, the principle of “like attracts like”. To study the properties of this model, we introduce two novel social network inter-connectivity measures, “clumpiness” and “hoprank,” that are the same basic concept defined at global and local levels, respectively. The measures may be computed from samples of readily available demographic data, and are useful for measuring probabilistic packet transmission through social networks. In three studies we apply clumpiness to measure the effects, on COVID-19 transmission, caused by social networks of both homophilic physical proximity and homophilic information replication. The particular characteristic we are interested in about disease transmission is herd immunity, the percentage of a population that has to be immune in order to prevent infection from spreading to those who are not. Two studies demonstrate innovations measuring herd immunity levels and predicting future outbreak locations, procedures relevant to epidemiological control policy. In the first study, we look at how homophilic physical proximity networks form natural bubbles that act as frictive surfaces that affect the speed of transmission of packets and influence herd immunity levels. In the second study, we test clumpiness in homophilic proximity social networks as a predictor of future infection outbreaks at the level of individual schools, restaurants, and workplaces. Our third study demonstrates that protective social bubbles form naturally from homophilic information replication networks, and enhance the natural bubbles that come from the homophilic physical proximity networks. Accurate description of this information environment lays the foundation for epidemiological messaging policy formation.

Pattern mining usually results in huge amounts of patterns, among which only small percentages are interesting. In this paper, Surprisingness (including Surpringness_I and Surpringness_II) is proposed as an innovative objective multivariate interestingness measure for automatically identifying interesting patterns from a large quantity of patterns. Surprisingness is applicable in unstructured or semi-structured, multi-domain or mixed-domain data compared to existing measures. An experiment has been conducted enabling unsupervised learning of common sense, interesting patterns and exceptions from a knowledge graph database built from Wikipedia <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> extracted data (represented as directed labeled hypergraphs), using Surpringness.

We present the cognitive architecture of an autonomous agent for active portfolio management in decentralized finance, involving activities such as asset selection, portfolio balancing, liquidity provision and trading. Partial implementation of the architecture is provided and supplied with preliminary results and conclusions.

Current climate simulation models provide valuable insights but are highly complicated, with numerous parameters, making them complex for assessing the causal impact of anthropogenic and natural factors on global temperatures. We applied multivariate Granger causality to investigate how combinations of forcings affect Earth's surface and ocean temperatures. Clear causal impact was found due to combinations of human factors including well-mixed greenhouse gases, tropical aerosols, ozone, and land use, while non-anthropogenic factors showed a non-trivial but lesser role.

It is argued that a fuzzy version of 4-truth-valued paraconsistent logic (with truth values corresponding to True, False, Both and Neither) can be approximately isomorphically mapped into the complex-number algebra of quantum probabilities. I.e., p-bits (paraconsistent bits) can be transformed into close approximations of qubits. The approximation error can be made arbitrarily small, at least in a formal sense, and can be related to the degree of irreducible "evidential error" assumed to plague an observer's observations. This logical correspondence manifests itself in program space via an approximate mapping between probabilistic and quantum types in programming languages.

It is argued that a broad class of AGI-relevant algorithms can be expressed in a common formal framework, via specifying Galois connections linking search and optimization processes on directed metagraphs whose edge targets are labeled with probabilistic dependent types, and then showing these connections are fulfilled by processes involving metagraph chronomorphisms. Examples are drawn from the core cognitive algorithms used in the OpenCog AGI framework: Probabilistic logical inference, evolutionary program learning, pattern mining, agglomerative clustering, pattern mining and nonlinear-dynamical attention allocation. The analysis presented involves representing these cognitive algorithms as recursive discrete decision processes involving optimizing functions defined over metagraphs, in which the key decisions involve sampling from probability distributions over metagraphs and enacting sets of combinatory operations on selected sub-metagraphs. The mutual associativity of the combinatory operations involved in a cognitive process is shown to often play a key role in enabling the decomposition of the process into folding and unfolding operations; a conclusion that has some practical implications for the particulars of cognitive processes, e.g. militating toward use of reversible logic and reversible program execution. It is also observed that where this mutual associativity holds, there is an alignment between the hierarchy of subgoals used in recursive decision process execution and a hierarchy of subpatterns definable in terms of formal pattern theory.

A novel optimization strategy, Info-Evo, is described, in which natural gradient search using nonparametric Fisher information is used to provide ongoing guidance to an evolutionary learning algorithm, so that the evolutionary process preferentially moves in the directions identified as "shortest paths" according to the natural gradient. Some specifics regarding the application of this approach to automated program learning are reviewed, including a strategy for integrating Info-Evo into the MOSES program learning framework.

A multi-decade exploration into the theoretical foundations of artificial and natural general intelligence, which has been expressed in a series of books and papers and used to guide a series of practical and research-prototype software systems, is reviewed at a moderate level of detail. The review covers underlying philosophies (patternist philosophy of mind, foundational phenomenological and logical ontology), formalizations of the concept of intelligence, and a proposed high level architecture for AGI systems partly driven by these formalizations and philosophies. The implementation of specific cognitive processes such as logical reasoning, program learning, clustering and attention allocation in the context and language of this high level architecture is considered, as is the importance of a common (e.g. typed metagraph based) knowledge representation for enabling "cognitive synergy" between the various processes. The specifics of human-like cognitive architecture are presented as manifestations of these general principles, and key aspects of machine consciousness and machine ethics are also treated in this context. Lessons for practical implementation of advanced AGI in frameworks such as OpenCog Hyperon are briefly considered.

MeTTa (Meta Type Talk) is a novel programming language created for use in the OpenCog Hyperon AGI system. It is designed as a meta-language with very basic and general facilities for handling symbols, groundings, variables, types, substitutions and pattern matching. Primitives exist for creating new type systems and associated DSLs. IInformally, MeTTa is Hyperon's lowest-level "language of thought" -- the meta-language in which algorithms for learning more particular knowledge representations, will operate, and in which these algorithms themselves may be represented. Here we explain how one might go about formalizing the MeTTa language as a system of metagraph rewrite rules, an approach that fits in naturally to the Hyperon framework given that the latter's core component is a distributed metagraph knowledge store (the Atomspace). The metagraph rewrite rules constituting MeTTa programs can also be represented as metagraphs, giving a natural model for MeTTa reflection and self-modifying code. Considering MeTTa programs that compute equivalences between execution traces of other MeTTa programs allows us to model spaces of MeTTa execution traces using Homotopy Type Theory. Considering the limit of MeTTa programs mapping between execution traces of MeTTa programs that map between execution traces of MeTTa programs that ..., we find that a given MeTTa codebase is effectively modeled as an infinity-groupoid, and the space of all MeTTa codebases as an (infinity,1)-topis This topos is basically the same as the so-called "Ruliad" previously derived from rewrite rules on hypergraphs, in a discrete physics context. The formalization of MeTTA as metagraph rewrite rules may also provide a useful framework for structuring the implementation of efficient methods for pattern matching and equality inference within the MeTTa interpreter.

What kind of programming language would be most appropriate to serve the needs of integrative, multi-paradigm, multi-software-system approaches to AGI? This question is broached via exploring the more particular question of how to create a more scalable and usable version of the "Atomese" programming language that forms a key component of the OpenCog AGI design (an "Atomese 2.0") . It is tentatively proposed that the core of Atomese 2.0 should be a very flexible framework of rewriting rules for rewriting a metagraph (where the rules themselves are represented within the same metagraph, and some of the intermediate data created and used during the rule-interpretation process may be represented in the same metagraph). This framework should support concurrent rewriting of the metagraph according to rules that are labeled with various sorts of uncertainty-quantifications, and that are labeled with various sorts of types associated with various type systems. A gradual typing approach should be used to enable mixture of rules and other metagraph nodes/links associated with various type systems, and untyped metagraph nodes/links not associated with any type system. This must be done in a way that allows reasonable efficiency and scalability, including in concurrent and distributed processing contexts, in the case where a large percentage of of processing time is occupied with evaluating static pattern-matching queries on specific subgraphs of a large metagraph (including a rich variety of queries such as matches against nodes representing variables, and matches against whole subgraphs, etc.).

A novel computational model (CoDD) utilizing combinatory logic to create higher-order decision trees is presented. A theoretical analysis of general intelligence in terms of the formal theory of pattern recognition and pattern formation is outlined, and shown to take especially natural form in the case where patterns are expressed in CoDD language. Relationships between logical entropy and algorithmic information, and Shannon entropy and runtime complexity, are shown to be elucidated by this approach. Extension to the quantum computing case is also briefly discussed.

A formal theory of simplicity is introduced, in the context of a "combinational" computation model that views computation as comprising the iterated transformational and compositional activity of a population of agents upon each other. Conventional measures of simplicity in terms of algorithmic information etc. are shown to be special cases of a broader understanding of the core "symmetry" properties constituting what is defined here as a Compositional Simplicity Measure (CoSM). This theory of CoSMs is extended to a theory of CoSMOS (Combinational Simplicity Measure Operating Sets) which involve multiple simplicity measures utilized together. Given a vector of simplicity measures, an entity is associated not with an individual simplicity value but with a "simplicity bundles" of Pareto-optimal simplicity-value vectors. CoSMs and CoSMOS are then used as a foundation for a theory of pattern and multipattern, and a theory of hierarchy and heterarchy in systems of patterns. A formalization of the cognitive-systems notion of a "coherent dual network" interweaving hierarchy and heterarchy in a consistent way is presented. The high level end result of this investigation is to re-envision Occam's Razor as something like: When in doubt, prefer hypotheses whose simplicity bundles are Pareto optimal, partly because doing so both permits and benefits from the construction of coherent dual networks comprising coordinated and consistent multipattern hierarchies and heterarchies.

Ideas and formalisms from far-from-equilibrium thermodynamics are ported to the context of stochastic computational processes, via following and extending Tadaki's algorithmic thermodynamics. A Principle of Maximum Algorithmic Caliber is proposed, providing guidance as to what computational processes one should hypothesize if one is provided constraints to work within. It is conjectured that, under suitable assumptions, computational processes obeying algorithmic Markov conditions will maximize algorithmic caliber. It is proposed that in accordance with this, real-world cognitive systems may operate in substantial part by modeling their environments and choosing their actions to be (approximate and compactly represented) algorithmic Markov networks. These ideas are suggested as potential early steps toward a general theory of the operation of pragmatic generally intelligent systems.

A novel approach to automated learning of syntactic rules governing natural languages is proposed, based on using probabilities assigned to sentences (and potentially longer word sequences) by transformer neural network language models to guide symbolic learning processes like clustering and rule induction. This method exploits the learned linguistic knowledge in transformers, without any reference to their inner representations; hence, the technique is readily adaptable to the continuous appearance of more powerful language models. We show a proof-of-concept example of our proposed technique, using it to guide unsupervised symbolic link-grammar induction methods drawn from our prior research.

The DeepWalk algorithm is used to assign embedding vectors to nodes in the Atomspace weighted, labeled hypergraph that is used to represent knowledge in the OpenCog AGI system, in the context of an application to probabilistic inference regarding the causes of longevity based on data from biological ontologies and genomic analyses. It is shown that vector difference operations between embedding vectors are, in appropriate conditions, approximately alignable with "intensional difference" operations between the hypergraph nodes corresponding to the embedding vectors. This relationship hints at a broader functorial mapping between uncertain intensional logic and vector arithmetic, and opens the door for using embedding vector algebra to guide intensional inference control.

Beginning with a simple semantics for propositions, based on counting observations, it is shown that probabilistic and fuzzy logic correspond to two different heuristic assumptions regarding the combination of propositions whose evidence bases are not currently available. These two different heuristic assumptions lead to two different sets of formulas for propagating quantitative truth values through lattice operations. It is shown that these two sets of formulas provide a natural grounding for the multiplicative and additive operator-sets in linear logic. The standard rules of linear logic then emerge as consequences of the underlying semantics. The concept of linear logic as a ``logic of resources" is manifested here via the principle of ``conservation of evidence" -- the restrictions to weakening and contraction in linear logic serve to avoid double-counting of evidence (beyond any double-counting incurred via use of heuristic truth value functions).

Typed metagraphs are defined as hypergraphs with types assigned to hyperedges and their targets, and the potential to have targets of hyperedges connect to whole links as well as targets. Directed typed metagraphs (DTMGs) are introduced via partitioning the targets of each edge in a typed metagraph into input, output and lateral sets; one can then look at "metapaths" in which edges' output-sets are linked to other edges' input-sets. An initial algebra approach to DTMGs is presented, including introduction of constructors for building up DTMGs and laws regarding relationships among multiple ways of using these constructors. A menagerie of useful morphism types is then defined on DTMGs (catamorphisms, anamorphisms, histomorphisms, futumorphisms, hylomorphisms, chronomorphisms, metamorphisms and metachronomorphisms), providing a general abstract framework for formulating a broad variety of metagraph operations. Deterministic and stochastic processes on typed metagraphs are represented in terms of forests of DTMGs defined over a common TMG, where the various morphisms can be straightforwardly extended to these forests. A variation of the approach to undirected typed metagraphs is presented; and it is indicated how the framework outlined can applied to realistic metagraphs involving complexities like dependent and probabilistic types, multidimensional values and dynamic processing including insertion and deletion of edges.

It is argued that 4-valued paraconsistent truth values (called here "p-bits") can serve as a conceptual, mathematical and practical foundation for highly AI-relevant forms of probabilistic logic and probabilistic programming and concept formation. First it is shown that appropriate averaging-across-situations and renormalization of 4-valued p-bits operating in accordance with Constructible Duality (CD) logic yields PLN (Probabilistic Logic Networks) strength-and-confidence truth values. Then variations on the Curry-Howard correspondence are used to map these paraconsistent and probabilistic logics into probabilistic types suitable for use within dependent type based programming languages. Zach Weber's paraconsistent analysis of the sorites paradox is extended to form a paraconsistent / probabilistic / fuzzy analysis of concept boundaries; and a paraconsistent version of concept formation via Formal Concept Analysis is presented, building on a definition of fuzzy property-value degrees in terms of relative entropy on paraconsistent probability distributions. These general points are fleshed out via reference to the realization of probabilistic reasoning and programming and concept formation in the OpenCog AGI framework which is centered on collaborative multi-algorithm updating of a common knowledge metagraph.

A new variation of the Kendall correlation, the "Probabilistic Support Kendall Correlation" (PSKC), is proposed based on applying the notion of “probabilistic support” to compare the pairwise comparisons of measurements. It is shown that the most basic version of the PSKC is proportional to the standard Kendall correlation under the assumption of no ties; however, the PSKC also lends itself to various extensions involving restrictions to specific sorts of comparisons or consideration of the relative magnitudes of different comparisons (the latter being the PSCC or Probabilistic Support Comparison Correlation as introduced here). It is shown that under broad conditions, Probabilistic Support Kendall Correlation (and hence the standard Kendall correlation as well as the various more general versions of PSKC) has a strong, elegant transitivity property.

Ideas and formalisms from far-from-equilibrium thermodynamics are ported to the context of stochastic computational processes, via following and extending Tadaki's algorithmic thermodynamics. A Principle of Maximum Algorithmic Caliber is proposed, providing guidance as to what computational processes one should hypothesize if one is provided constraints to work within. It is conjectured that, under suitable assumptions, computational processes obeying algorithmic Markov conditions will maximize algorithmic caliber. It is proposed that in accordance with this, real-world cognitive systems may operate in substantial part by modeling their environments and choosing their actions to be (approximate and compactly represented) algorithmic Markov networks. These ideas are suggested as potential early steps toward a general theory of the operation of pragmatic generally intelligent systems.

The accelerating pace of development in artificial intelligence invokes fundamental questions regarding what constitutes a mind. Considering that AI has sprung from the cognitive enterprise of humanity and that there is the likely prospect of constructing some form of 'global brain', examining such questions may help to refine the conversation around AI and understand what might constitute a global brain. We draw on recent work in philosophy of mind and cognition—viewing mind as being representational and/or enactive in nature and function—, and hone in on insights that practices and accounts of non-ordinary consciousness (NOC) lend to this discussion. We discuss the concept of a "mindplex", which is comprised of individual constituents (humans and/or AI's) and an encompassing collective global brain that both concurrently possess cohesive 'theatres of consciousness'. We posit that these theatres of an AGI global brain have correspondences with key aspects of the inherent nature of consciousness and reality (as understood outside the AI context) as gleaned through NOC. We propose that these considerations are poised to strongly contribute to the construction of mindplexes by reflecting in their functional architecture the nature of mind and engineering mindplexes in tandem with ethical and value-alignment considerations. Furthermore, we propose the adoption and implementation of AI on distributed ledger technology (DLT), such as blockchain, which would provide support for the evolution of a global brain mindplex.

A new conceptual foundation for the notion of "information" is proposed, based on the concept of a "distinction graph": a graph in which two nodes are connected iff they cannot be distinguished by a particular observer. The "graphtropy" of a distinction graph is defined as the average connection probability of two nodes; in the case where the distinction graph is a composed of disconnected components that are fully connected subgraphs, this is equivalent to Ellerman's logical entropy, which has straightforward relationships to Shannon entropy. Probabilistic distinction graphs and probabilistic graphtropy are also considered, as well as connections between graphtropy and thermodynamic and quantum entropy. The semantics of the Second Law of Thermodynamics and the Maximum Entropy Production Principle are unfolded in a novel way, via analysis of the cognitive processes underlying the making of distinction graphs This evokes an interpretation in which complex intelligence is seen to correspond to states of consciousness with intermediate graphtropy, which are associated with memory imperfections that violate the assumptions leading to derivation of the Second Law. In the case where nodes of a distinction graph are labeled by computable entities, graphtropy is shown to be monotonically related to the average algorithmic information of the nodes (relative to to the algorithmic information of the observer). A quantum-mechanical version of distinction graphs is considered, in which distinctions can exist in a superposed state; this yields to graphtropy as a measure of the impurity of a mixed state, and to a concept of "quangraphtropy." Finally, a novel computational model called Dynamic Distinction Graphs (DDGs) is formulated, via enhancing distinction graphs with additional links expressing causal implications, enabling a distinction-based model of "observers."

In this work we explore the implementation of the reputation system for a generic marketplace, describe details of the algorithm and parameters driving its operation, justify an approach to simulation modeling, and explore how various kinds of reputation systems with different parameters impact the economic security of the marketplace. Our emphasis here is on the protection of consumers by means of an ability to distinguish between cheating participants and honest participants, as well as the ability to minimize losses of honest participants due to scam.

We present an exploration of a reputation system based on explicit ratings weighted by the values of corresponding financial transactions from the perspective of its ability to grant "security" to market participants by protecting them from scam and "equity" in terms of having real qualities of the participants correctly assessed. We present a simulation modeling approach based on the selected reputation system and discuss the results of the simulation.

Current brain imaging technology is not able to detect the dynamics of thoughts and feelings as they pass through the brain. Artificial Intelligence (AI) experts agree that this sort of Artificial General Intelligence (AGI) will eventually be possible, but estimates of the timeline and the most likely path to this goal differ widely. motion capture should be supplemented with emotion capture: a variety of physical sensors capturing parameters such as heart rate, galvanic skin response, and so forth, giving a physiological indication of the user's emotional state. There is significant ongoing work in this direction as well, though not specifically directed at virtual agent control. Robot simulators exceed virtual worlds and game engines in terms of allowing detailed control of simulated robot joints, and physically realistic interactions between objects (as needed for instance to enable tool use), but don't allow large numbers of individuals to log on and help teach robots.

One approach to achieving artificial general intelligence (AGI) is through the emergence of complex structures and dynamic properties arising from decentralized networks of interacting artificial intelligence (AI) agents. Understanding the principles of consensus in societies and finding ways to make consensus more reliable becomes critically important as connectivity and interaction speed increase in modern distributed systems of hybrid collective intelligences, which include both humans and computer systems. We propose a new form of reputation-based consensus with greater resistance to reputation gaming than current systems have. We discuss options for its implementation, and provide initial practical results.

Understanding the principles of consensus in communities and finding ways to optimal solutions beneficial for entire community becomes crucial as the speeds and scales of interaction in modern distributed systems increase. Such systems can be both social and information computer networks that unite the masses of people as well as multi-agent computing platforms based on peer-to-peer interactions including those operating on the basis of distributed ledgers. It is now becoming possible for hybrid ecosystems to emerge, having such systems including both humans and computer systems using artificial intelligence. We propose a new form of consensus for all of the listed systems, based on the reputation of the participants, calculated according to the principle of "liquid democracy". We believe that such a system will be more resistant to social engineering and reputation manipulation than the existing systems. In this article, we discuss the basic principles and options for implementing such a system, and also present preliminary practical results.

Background: Numerous reports on Alzheimer’s Disease (AD) indicate that AD has many biochemical pathways. However, most AD treatments have targeted a single pathway such as beta amyloid or phosphorylated tau and have not been very successful in stopping AD progression. To address the multipath nature of AD, we have used multiple natural products that target critical aging and independent AD pathways to test whether a multipath approach might be more effective than the single pathway approaches typically used. Methods and findings: We have constructed Drosophila melanogaster models of AD with inducible transgenic human Amyloid-β and tau mutations. Induction of either AD mutant gene in the transgenic flies after emergence from pupal stage leads to the early onset of slowed mobility that was tracked by assaying crawl times as a function of age. The transgenic Amyloid-β and tau Drosophila assays were utilized to test various natural products that target separate biological pathways involved in brain aging and dementia. Screening a library of natural products, we identified a group of 7 natural substances (MX100A) that synergistically prevented nearly all early mobility difficulties and reverses the lifeshortening effects of Amyloid-β and tau. The favorable results on AD symptoms in Drosophila with MX100A were observed in animals treated from youth and from later adult ages. MX100A had no observed negative side effects. Conclusion: Treatment of transgenic AD Drosophila with the multipath MX100A dietary supplement prevents nearly all early mobility difficulties and reverses the lifeshortening effects of Amyloid-β and tau. In patients with AD, progressively more severe mobility difficulties and heightened risks of mortality are common symptoms of the later stages of AD. While successful animal results have not typically translated into effective treatments for AD, the multipath MX100A treatment results are promising enough to warrant further testing.

Context: Currently, there is no treatment that can stop the progression of Alzheimer’s disease. We have used transgenic Drosophila melanogaster models and machine learning to develop an eight component botanical mixture (Geneaire™ ReBuilder™) that targets multiple genetic pathways involved in brain aging and dementia that are homologous between Drosophila and humans. Objective: To test the effects of ReBuilder on the cognitive function of subjects diagnosed with mild or moderate Alzheimer’s disease. Methods: We recruited 50 subjects with mild to moderate AD to participate in a double-blind, placebo-controlled clinical study. During the 12-month pilot study, the subjects were evaluated quarterly on the Mini Mental State Exam (MMSE), Alzheimer’s Disease Cooperative Study’s Activities of Daily Living (ADCS-ADL), and the Clinical Dementia Rating Sum of Boxes (CDR-SB). Results: The addition of ReBuilder to subjects’ existing Namenda and Exelon regimens stabilized cognitive decline in patients with mild AD and slowed cognitive decline in patients with moderate AD. Conclusions: These results were observed in both sexes and in all ages tested. Importantly, no adverse side effects attributable to ReBuilder were reported. The results of this clinical pilot warrant further study of ReBuilder in AD.

Understanding the principles of consensus in communities and finding ways to optimal solutions beneficial for entire community becomes crucial as the speeds and scales of interaction in modern distributed systems increase. Such systems can be both social and information computer networks that unite the masses of people as well as multi-agent computing platforms based on peer-to-peer interactions including those operating on the basis of distributed ledgers. It is now becoming possible for hybrid ecosystems to emerge, having such systems including both humans and computer systems using artificial intelligence. We propose a new form of consensus for all of the listed systems, based on the reputation of the participants, calculated according to the principle of liquid democracy. We believe that such a system will be more resistant to social engineering and reputation manipulation than the existing systems. In this article, we discuss the basic principles and options for implementing such a system, and also present preliminary practical results.

"Cognitive synergy" refers to a dynamic in which multiple cognitive processes, cooperating to control the same cognitive system, assist each other in overcoming bottlenecks encountered during their internal processing. Cognitive synergy has been posited as a key feature of real-world general intelligence, and has been used explicitly in the design of the OpenCog cognitive architecture. Here category theory and related concepts are used to give a formalization of the cognitive synergy concept. A series of formal models of intelligent agents is proposed, with increasing specificity and complexity: simple reinforcement learning agents; "cognit" agents with an abstract memory and processing model; hypergraph-based agents (in which "cognit" operations are carried out via hypergraphs); hypergraph agents with a rich language of nodes and hyperlinks (such as the OpenCog framework provides); "PGMC" agents whose rich hypergraphs are endowed with cognitive processes guided via Probabilistic Growth and Mining of Combinations; and finally variations of the PrimeAGI design, which is currently being built on top of OpenCog. A notion of cognitive synergy is developed for cognitive processes acting within PGMC agents, based on developing a formal notion of "stuckness," and defining synergy as a relationship between cognitive processes in which they can help each other out when they get stuck. It is proposed that cognitive processes relating to each other synergetically, associate in a certain way with functors that map into each other via natural transformations. Cognitive synergy is proposed to correspond to a certain inequality regarding the relative costs of different paths through certain commutation diagrams. Applications of this notion of cognitive synergy to particular cognitive phenomena, and specific cognitive processes in the PrimeAGI design, are discussed.

A new model of symbol grounding is presented, in which the structures of natural language, logical semantics, perception and action are represented categorically, and symbol grounding is modeled via the composition of morphisms between the relevant categories. This model gives conceptual insight into the fundamentally systematic nature of symbol grounding, and also connects naturally to practical real-world AI systems in current research and commercial use. Specifically, it is argued that the structure of linguistic syntax can be modeled as a certain asymmetric monoidal category, as e.g. implicit in the link grammar formalism; the structure of spatiotemporal relationships and action plans can be modeled similarly using "image grammars" and "action grammars"; and common-sense logical semantic structure can be modeled using dependently-typed lambda calculus with uncertain truth values. Given these formalisms, the grounding of linguistic descriptions in spatiotemporal perceptions and coordinated actions consists of following morphisms from language to logic through to spacetime and body (for comprehension), and vice versa (for generation). The mapping is indicated between the spatial relationships in the Region Connection Calculus and Allen Interval Algebra and corresponding entries in the link grammar syntax parsing dictionary. Further, the abstractions introduced here are shown to naturally model the structures and systems currently being deployed in the context of using the OpenCog cognitive architecture to control Hanson Robotics humanoid robots.

A novel partial order is defined on the space of digraphs or hypergraphs, based on assessing the cost of producing a graph via a sequence of elementary transformations. Leveraging work by Knuth and Skilling on the foundations of inference, and the structure of Heyting algebras on graph space, this partial order is used to construct an intuitionistic probability measure that applies to either digraphs or hypergraphs. As logical inference steps can be represented as transformations on hypergraphs representing logical statements, this also yields an intuitionistic probability measure on spaces of theorems. The central result is also extended to yield intuitionistic probabilities based on more general weighted rule systems defined over bicartesian closed categories.

A general formulation of optimization problems in which various candidate solutions may use different feature-sets is presented, encompassing supervised classification, automated program learning and other cases. A novel characterization of the concept of a "good quality feature" for such an optimization problem is provided; and a proposal regarding the integration of quality based feature selection into metalearning is suggested, wherein the quality of a feature for a problem is estimated using knowledge about related features in the context of related problems. Results are presented regarding extensive testing of this "feature metalearning" approach on supervised text classification problems; it is demonstrated that, in this context, feature metalearning can provide significant and sometimes dramatic speedup over standard feature selection heuristics.

The "Loving AI" project involves developing software enabling humanoid robots to interact with people in loving and compassionate ways, and to promote people' self-understanding and self-transcendence. Currently the project centers on the Hanson Robotics robot "Sophia" -- specifically, on supplying Sophia with personality content and cognitive, linguistic, perceptual and behavioral content aimed at enabling loving interactions supportive of human self-transcendence. In September 2017 a small pilot study was conducted, involving the Sophia robot leading human subjects through dialogues and exercises focused on meditation, visualization and relaxation. The pilot was an apparent success, qualitatively demonstrating the viability of the approach and the ability of appropriate human-robot interaction to increase human well-being and advance human consciousness.

Leslie Kean’s Surviving Death is a wonderfully readable, carefully constructed summary of the evidence for the existence of what is colloquially called an “afterlife.” That is, she considers evidence for the hypothesis that individual human minds and personalities possess an existence going beyond their attachment to any particular body – so that, for instance, an individual with a certain name and certain traits may sometimes continue to perceive and act, even when the body typically associated with that individual is dead and gone. Most of the book comprises moderately detailed descriptions of specific cases, involving specific people, which indicate the existence of some sort of “afterlife” for individual human minds, or potentially give some information regarding the nature of this afterlife. Kean considers a gamut of phenomena such as past-life memories, near-death experiences, mediumistic trances, poltergeists, and so forth. However, she also makes a significant effort to draw general conclusions, lessons and hypotheses from the totality of these cases, while maintaining respect for the confusing and in many ways still mysterious nature of the phenomena under discussion. Each of the topics considered in the book has been reviewed and analyzed in more depth elsewhere. What Kean does, however, is provide a clear, evocative and rational survey of the many types of evidence that are directly relevant to the possibility and nature of an afterlife for individual human minds.

Two theses are proposed, regarding the future evolution of the value systems of advanced AGI systems. The Value Learning Thesis is a semi-formalized version of the idea that, if an AGI system is taught human values in an interactive and experiential way as its intelligence increases toward human level, it will likely adopt these human values in a genuine way. The Value Evolution Thesis is a semi-formalized version of the idea that if an AGI system begins with human-like values, and then iteratively modifies itself, it will end up in roughly the same future states as a population of human beings engaged with progressively increasing their own intelligence (e.g. by cyborgification or brain modification). Taken together, these theses suggest a worldview in which raising young AGIs to have human-like values is a sensible thing to do, and likely to produce a future that is generally desirable in a human sense.&#x0D; While these two theses are far from definitively proven, I argue that they are more solid and more relevant to the actual future of AGI than Bostrom’s “Instrumental Convergence Thesis” and “Orthogonality Thesis” which are core to the basis of his argument (in his book Superintelligence) for fearing ongoing AGI development and placing AGI R&amp;D under strict governmental control. In the context of fleshing out this argument, previous publications and discussions by Richard Loosemore and Kaj Sotala are discussed in some detail.

This research describes a novel approach to incorporating navigation in a general planning process. Five planning control strategies are articulated for combining "primitive rules" (which directly solve problems) with "recursive rules" (which decompose problems into combinations of problems of the same form). The combined rules are applied to carry out navigation and generalized planning, utilizing the observation that a "planning-aided navigation" problem can be decomposed automatically into smaller planning-aided navigation problems, and hence addressed by vertical recursive rules. The approach is illustrated in a MineCraft construction domain, using an agent that is able to build blocks to create a path to reach an initially unreachable target, e.g., a path finding task in which parts of the path must be constructed during the course of navigation/planning, such as building a bridge to fill a gap on the way to the target location.

A high-level artificial general intelligence (AGI) architecture called goal-oriented learning meta-architecture (GOLEM) is presented, along with an informal but careful argument that GOLEM may be capable of preserving its initial goals while radically improving its general intelligence. As a meta-architecture, GOLEM can be wrapped around a variety of different base-level AGI systems, and also has a role for a powerful narrow-AI subcomponent as a probability estimator. The motivation underlying these ideas is the desire to create AGI systems fulfilling the multiple criteria of being massively and self-improvingly intelligent, probably beneficial, and almost surely not destructive.

This edited volume gives the first-ever book-length presentation of contemporary research in the domain of Artificial General Intelligence (AGI). What distinguishes AGI research from run-of-the-mill AI research is that it is explicitly focused on engineering general intelligence ? autonomous, self-reflective, self-improving, commonsensical intelligence ? in the short term. Bringing the diverse body of AGI research together in a single volume reveals the common themes among various researchers? work, and makes clear what the big open questions are in this vital and critical area of research, as well as highlighting relationships between AI and related fields such as philosophy, neuroscience, linguistics, psychology, biology, sociology, anthropology and engineering.

Oxford philosopher Nick Bostrom, in his recent and celebrated book Superintelligence, argues that advanced AI poses a potentially major existential risk to humanity, and that advanced AI development should be heavily regulated and perhaps even restricted to a small set of government-approved researchers. Bostrom’s ideas and arguments are reviewed and explored in detail, and compared with the thinking of three other current thinkers on the nature and implications of AI: Eliezer Yudkowsky of the Machine Intelligence Research Institute (formerly Singularity Institute for AI), and David Weinbaum (Weaver) and Viktoras Veitas of the Global Brain Institute. Relevant portions of Yudkowsky’s book Rationality: From AI to Zombies are briefly reviewed, and it is found that nearly all the core ideas of Bostrom’s work appeared previously or concurrently in Yudkowsky’s thinking. However, Yudkowsky often presents these shared ideas in a more plain-spoken and extreme form, making clearer the essence of what is being claimed. For instance, the elitist strain of thinking that one sees in the background in Bostrom is plainly and openly articulated in Yudkowsky, with many of the same practical conclusions (e.g. that it may well be best if advanced AI is developed in secret by a small elite group). Bostrom and Yudkowsky view intelligent systems through the lens of reinforcement learning – they view them as “reward-maximizers” and worry about what happens when a very powerful and intelligent reward-maximizer is paired with a goal system that gives rewards for achieving foolish goals like tiling the universe with paperclips. Weinbaum and Veitas’s recent paper “Open-Ended Intelligence” presents a starkly alternative perspective on intelligence, viewing it as centered not on reward maximization, but rather on complex self-organization and self-transcending development that occurs in close coupling with a complex environment that is also ongoingly self-organizing, in only partially knowable ways. It is concluded that Bostrom and Yudkowsky’s arguments for existential risk have some logical foundation, but are often presented in an exaggerated way. For instance, formal arguments whose implication is that the “worst case scenarios” for advanced AI development are extremely dire, are often informally discussed as if they demonstrated the likelihood, rather than just the possibility, of highly negative outcomes. And potential dangers of reward-maximizing AI are taken as problems with AI in general, rather than just as problems of the reward-maximization paradigm as an approach to building superintelligence. If one views past, current, and future intelligence as “open-ended,” in the vernacular of Weaver and Veitas, the potential dangers no longer appear to loom so large, and one sees a future that is wide-open, complex and uncertain, just as it has always been.

Oxford philosopher Nick Bostrom, in his recent and celebrated book Superintelligence, argues that advanced AI poses a potentially major existential risk to humanity, and that advanced AI development should be heavily regulated and perhaps even restricted to a small set of government-approved researchers. Bostrom’s ideas and arguments are reviewed and explored in detail, and compared with the thinking of three other current thinkers on the nature and implications of AI: Eliezer Yudkowsky of the Machine Intelligence Research Institute (formerly Singularity Institute for AI), and David Weinbaum (Weaver) and Viktoras Veitas of the Global Brain Institute.   Relevant portions of Yudkowsky’s book Rationality: From AI to Zombies are briefly reviewed, and it is found that nearly all the core ideas of Bostrom’s work appeared previously or concurrently in Yudkowsky’s thinking. However, Yudkowsky often presents these shared ideas in a more plain-spoken and extreme form, making clearer the essence of what is being claimed. For instance, the elitist strain of thinking that one sees in the background in Bostrom is plainly and openly articulated in Yudkowsky, with many of the same practical conclusions (e.g. that it may well be best if advanced AI is developed in secret by a small elite group).   Bostrom and Yudkowsky view intelligent systems through the lens of reinforcement learning – they view them as “reward-maximizers” and worry about what happens when a very powerful and intelligent reward-maximizer is paired with a goal system that gives rewards for achieving foolish goals like tiling the universe with paperclips. Weinbaum and Veitas’s recent paper “Open-Ended Intelligence” presents a starkly alternative perspective on intelligence, viewing it as centered not on reward maximization, but rather on complex self-organization and self-transcending development that occurs in close coupling with a complex environment that is also ongoingly self-organizing, in only partially knowable ways.   It is concluded that Bostrom and Yudkowsky’s arguments for existential risk have some logical foundation, but are often presented in an exaggerated way. For instance, formal arguments whose implication is that the “worst case scenarios” for advanced AI development are extremely dire, are often informally discussed as if they demonstrated the likelihood, rather than just the possibility, of highly negative outcomes. And potential dangers of reward-maximizing AI are taken as problems with AI in general, rather than just as problems of the reward-maximization paradigm as an approach to building superintelligence. If one views past, current, and future intelligence as “open-ended,” in the vernacular of Weaver and Veitas, the potential dangers no longer appear to loom so large, and one sees a future that is wide-open, complex and uncertain, just as it has always been.

Synthesizing concepts and findings from a number of recent models of human consciousness, a unified model of the key properties characterizing human consciousness is outlined. Six key properties are emphasized: Dynamical representation of the focus of consciousness, Focusing of energetic resources and focusing of informational resources on a subset of system knowledge, Global Workspace dynamics as outlined by Bernard Baars in his cognitive theory of consciousness, Integrated Information as emphasized by Tononi, and correlation of attentional focus with self-modeling. It is proposed that the extent, and relative importance, of these properties may vary in different states of consciousness; and that any AI system displaying closely human-like intelligence will need to manifest these properties in its consciousness as well. The “hard problem” of consciousness is sidestepped throughout, via focusing on structures and dynamics posited to serve as neural or cognitive correlates of subjective conscious experience.

This paper summarizes the authors’ thinking regarding the design of a software framework for interfacing between general-intelligence-oriented software systems and complex mobile robots, including humanoid robots. The framework describes incorporates perception synthesis, action orchestration, and high level control, and is designed to effectively leverage existing relevant software frameworks such as ROS and Blender. An initial case study motivating this work is the use of the OpenCog AGI (Artificial General Intelligence) software framework to help control humanoid robots created by Hanson Robotics.

A high-level artificial general intelligence (AGI) architecture called goal-oriented learning meta-architecture (GOLEM) is presented, along with an informal but careful argument that GOLEM may be capable of preserving its initial goals while radically improving its general intelligence. As a meta-architecture, GOLEM can be wrapped around a variety of different base-level AGI systems, and also has a role for a powerful narrow-AI subcomponent as a probability estimator. The motivation underlying these ideas is the desire to create AGI systems fulfilling the multiple criteria of being: massively and self-improvingly intelligent, probably beneficial and almost surely not destructive.

A novel approach to uncertain temporal inference is presented. Allen Interval Algebra is extended to fuzzy time-intervals via representing the latter as trapeziums with distinct beginning, middle and end. An uncertain version of the Interval Algebra composition table is developed via running a computer simulation in which a large number of fuzzy time-intervals are drawn from an assumed probability distribution.

A novel category of theories is proposed, providing a potential explanation for the representation of complex knowledge in the human (and, more generally, mammalian) brain. Firstly, a "glocal" representation for concepts is suggested, involving localized representations in a sparse network of "concept neurons" in the Medial Temporal Lobe, coupled with a complex dynamical attractor representation in other parts of cortex. Secondly, it is hypothesized that a combi-natory logic like representation is used to encode abstract relationships without explicit use of variable bindings, perhaps using systematic asynchronization among concept neurons to indicate an analogue of the combinatory-logic operation of function application. While unraveling the specifics of the brain's knowledge representation mechanisms will require data beyond what is currently available, the approach presented here provides a class of possibilities that is neurally plausible and bridges the gap between neurophysiological realities and mathematical and computer science concepts.

The Deep Spatio-Temporal Inference Network (DeSTIN) is a deep learning architecture which combines un-supervised learning and Bayesian inference. The original version of DeSTIN incorporates k-means clustering inside each processing node. Here we propose to replace k-means with a more sophisticated algorithm, online EM (Expectation Maximization), and show that this improves DeSTIN's performance on image classification and restoration tasks.

A practical roadmap to human-level artificial general intelligence is outlined, leading from the current situation in which complex AGI-oriented cognitive architectures remain partially implemented and inadequately tested, to a future in which AGI systems are deployed to carry out a variety of practical tasks currently only achievable via humans. The roadmap involves simultaneous development of proto-AGI applications in a number of different areas: videogame agent control, mobile robotics, genomics, natural language dialogue, theorem proving and program learning. The simultaneous pursuit of diverse application areas using a single AGI architecture is proposed as critical to maintaining generality of intelligence throughout the development process. While acknowledging the profound difficulty of time estimation for a project of this complexity, it is projected that the roadmap outlined could potentially be completed in a decade or less.

Abstract In recent years broad community of researchers has emerged, focusing on the original ambitious goals of the AI field - the creation and study of software or hardware systems with general intelligence comparable to, and ultimately perhaps greater than, that of human beings. This paper surveys this diverse community and its progress. Approaches to defining the concept of Artificial General Intelligence (AGI) are reviewed including mathematical formalisms, engineering, and biology inspired perspectives. The spectrum of designs for AGI systems includes systems with symbolic, emergentist, hybrid and universalist characteristics. Metrics for general intelligence are evaluated, with a conclusion that, although metrics for assessing the achievement of human-level AGI may be relatively straightforward (e.g. the Turing Test, or a robot that can graduate from elementary school or university), metrics for assessing partial progress remain more controversial and problematic.

The work outlines a novel conceptual and theoretical framework for understanding Artificial General Intelligence and based on this framework outlines a practical roadmap for the development of AGI with capability at the human level and ultimately beyond.

The work outlines a novel conceptual and theoretical framework for understanding Artificial General Intelligence and based on this framework outlines a practical roadmap for the development of AGI wit

Our goal in this book is straightforward, albeit ambitious: to present a conceptual and technical design for a thinking machine, a software program capable of the same qualitative sort of general intelligence as human beings.

Intelligence is not just about what happens inside a system, but also about what happens outside that system, and how the system interacts with its environment. Real-world general intelligence is about intelligence relative to some particular class of environments, and human-like general intelligence is about intelligence relative to the particular class of environments that humans evolved in (which in recent millennia has included environments humans have created using their intelligence). In Chap. 4 , we reviewed some specific capabilities characterizing human-like general intelligence; to connect these with the general theory of general intelligence from the last few chapters, we need to explain what aspects of human-relevant environments correspond to these human-like intelligent capabilities. We begin with aspects of the environment related to communication, which turn out to tie in closely with cognitive synergy. Then we turn to physical aspects of the environment, which we suspect also connect closely with various human cognitive capabilities. Finally we turn to physical aspects of the human body and their relevance to the human mind. In the following chapter we present a deeper, more abstract theoretical framework encompassing these ideas.

Real-world minds are always adapted to certain classes of environments and goals. The ideas of the previous chapter, regarding the connection between a human-like intelligence’s internals and its environment, result from exploring the implications of this adaptation in the context of the cognitive synergy concept. In this chapter we explore the mind-world connection in a broader and more abstract way – making a more ambitious attempt to move toward a “general theory of general intelligence”.

Creating AGI, we have said, is not only about having the right structural and dynamical possibilities implemented in the initial version of one’s system—but also about the environment and embodiment that one’s system is associated with, and the match between the system’s internals and these externals. Another key aspect is the long-term time-course of the system’s evolution over time, both in its internals and its external interaction—i.e., what is known as development.

Most commonly, if a work on advanced AI mentions ethics at all, it occurs in a final summary chapter, discussing in broad terms some of the possible implications of the technical ideas presented beforehand.

One of the most powerful metaphors we’ve found for understanding minds is to view them as networks—i.e. collections of interrelated, interconnected elements. The view of mind as network is implicit in the patternist philosophy, because every pattern can be viewed as a pattern in something, or a pattern of arrangement of something—thus a pattern is always viewable as a relation between two or more things. A collection of patterns is thus a pattern-network. Knowledge of all kinds may be given network representations; and cognitive processes may be represented as networks also; for instance via representing them as programs, which may be represented as trees or graphs in various standard ways. The emergent patterns arising in an intelligence as it develops may be viewed as a pattern network in themselves; and the relations between an embodied mind and its physical and social environment may be viewed in terms of ecological and social networks.

Explicit knowledge is easy to write about and talk about; implicit knowledge is equally important, but tends to get less attention in discussions of AI and psychology, simply because we don’t have as good a vocabulary for describing it, nor as good a collection of methods for measuring it. One way to deal with this problem is to describe implicit knowledge using language and methods typically reserved for explicit knowledge. This might seem intrinsically non-workable, but we argue that it actually makes a lot of sense. The same sort of networks that a system like CogPrime uses to represent knowledge explicitly, can also be used to represent the emergent knowledge that implicitly exists in an intelligent system’s complex structures and dynamics.

When one is involved with engineering an AGI system, one thinks a lot about the aspects of the system one is explicitly building—that are the parts, how they fit together, how to test they’re properly working, and so forth.

In conversations with government funding sources or narrow AI researchers about AGI work, one of the topics that comes up most often is that of “evaluation and metrics”.

Supposing the CogPrime approach to creating advanced AGI is workable—then what are the right practical steps to follow? The various structures and algorithms outlined in Part 2, should be engineered and software-tested, of course—but that’s only part of the study.

In the previous chapter we presented a roadmap aimed at taking AGI systems to human-level intelligence. But we also emphasized that the human level is not necessarily the upper limit. Indeed, it would be surprising if human beings happened to represent the maximal level of general intelligence possible, even with respect to the environments in which humans evolved.

Just as there are many different approaches to human flight—airplanes, helicopters, balloons, spacecraft, and doubtless many methods no person has thought of yet—similarly, there are likely many different approaches to advanced artificial general intelligence.

AGI design necessarily leads one into some rather abstract spaces—but being a human-like intelligence in the everyday world is a pretty concrete thing. If the CogPrime research program is successful, it will result not just in abstract ideas and equations, but rather in real AGI robots carrying out tasks in the world, and AGI agents in virtual worlds and online digital spaces conducting important business, doing science, entertaining and being entertained by us, and so forth.

CogPrime, the AGI architecture on which the bulk of this book focuses, is aimed at the creation of artificial general intelligence that is vaguely human-like in nature, and possesses capabilities at the human level and ultimately beyond.

In the last chapter we discussed human intelligence from a fairly down-to-earth perspective, looking at the particular intelligent functions that human beings carry out in their everyday lives. And we strongly feel this practical perspective is important: Without this concreteness, it’s too easy for AGI research to get distracted by appealing (or frightening) abstractions of various sorts.

While we believe CogPrime is the most thorough attempt at an architecture for advanced AGI, to date, we certainly recognize there have been many valuable attempts in the past with similar aims; and we also have great respect for other AGI efforts occurring in parallel with CogPrime development, based on alternative, sometimes overlapping, theoretical presuppositions and practical choices. In most of this book we will ignore these other current and historical efforts except where they are directly useful for CogPrime—there are many literature reviews already published, and this is a research treatise not a textbook. In this chapter, however, we will break from this pattern and give a rough high-level overview of the various AGI architectures at play in the field today. The overview definitely has a bias toward other work with some direct relevance to CogPrime, but not an overwhelming bias; we also discuss a number of approaches that are unrelated to, and even in some cases conceptually orthogonal to, our own.

When writing the first draft of this book, some years ago, we had the idea to explain CogPrime by aligning its various structures and processes with the ones in the “standard architecture diagram” of the human mind.

The artificial intelligence field is full of sophisticated mathematical models and equations, but most of these are highly specialized in nature—e.g. formalizations of particular logic systems, analyzes of the dynamics of specific sorts of neural nets, etc. On the other hand, a number of highly general models of intelligent systems also exist, including Hutter’s recent formalization of universal intelligence [Hut05] and a large body of work in the disciplines of systems science and cybernetics—but these have tended not to yield many specific lessons useful for engineering AGI systems, serving more as conceptual models in mathematical form.

As we have seen, the formal theory of general intelligence, in its current form, doesn’t really tell us much that’s of use for creating real-world AGI systems. It tells us that creating extraordinarily powerful general intelligence is almost trivial if one has unrealistically huge amounts of computational resources; and that creating moderately powerful general intelligence using feasible computational resources is all about creating AI algorithms and data structures that (explicitly or implicitly) match the restrictions implied by a certain class of situations, to which the general intelligence is biased.

The work outlines a detailed blueprint for the creation of an Artificial General Intelligence system with capability at the human level and ultimately beyond, according to the Cog Prime AGI design and the Open Cog software architecture.

The work outlines a detailed blueprint for the creation of an Artificial General Intelligence system with capability at the human level and ultimately beyond, according to the Cog Prime AGI design and

The primary burden of this book is to explain the CogPrime architecture for AGI—the broad outline of the design, the main dynamics it’s intended to display once complete, and the reasons why we believe it will be capable of leading to general intelligence at the human level and beyond.

In this chapter and the next we describe one particular incarnation of the above ideas on semantic CSDLNs in more depth: the integration of CogPrime with the DeSTIN architecture reviewed in Chap. 5 of Part 1.

While it’s widely accepted that human beings carry out both symbolic and subsymbolic processing, as integral parts of their general intelligence, the precise definition of “symbolic” versus “subsymbolic” is a subtle issue, which different AI researchers will approach in different ways depending on their differing overall perspectives on AI.

Broadly speaking, the learning of predicates and schemata (executable procedures) is done in CogPrime via a number of different methods, including for example PLN inference and concept predicatization (to be discussed in later chapters). Most of these methods, however, merely extrapolate procedures directly from other procedures or concepts in the AtomSpace, in a local way—a new procedure is derived from a small number of other procedures or concepts.

In procedure learning as elsewhere in cognition, it’s not enough to use the right algorithm, one has to use it in the right way based on the data and context and affordances available.

Having chosen to represent procedures as programs, explicit procedure learning then becomes a matter of automated program learning.

A program evolution component is proposed for integrative artificial general intelligence. The system’s deployment is intended to be comparable, on Marr’s level of computational theory, to evolutionary mechanisms in human thought. The challenges of program evolution are described, along with the requirements for a program evolution system to be competent- solving hard problems quickly, accurately, and reliably. Meta-optimizing semantic evolutionary search (MOSES) is proposed to fulfill these requirements.

Now we turn to CogPrime’s methods for handling declarative knowledge—beginning with a series of chapters discussing the Probabilistic Logic Networks (PLN) [GMIH08] approach to uncertain logical reasoning, and then turning to chapters on pattern mining and concept creation.

Spatiotemporal reasoning is an important skill that an AGI is expected to have, innately or not. Much work has already been done in defining reasoning systems for space, time and spacetime, such as the Region Connection Calculus for space, Allen’s Interval Algebra for time, or the Qualitative Trajectory Calculus for motion. However, these reasoning systems rarely take adequate account of uncertainty, which poses an obstacle to using them in an AGI system confronted with an uncertain reality. In this paper we show how to use PLN (Probabilistic Logic Networks) to represent spatiotemporal knowledge and reasoning, via incorporating existing spatiotemporal calculi, and considering a novel extension of standard PLN truth values inspired by $$\mathcal {P}(\mathcal {Z})$$ -logic. This “PLN-ization” of existing spatiotemporal calculi, we suggest, constitutes an approach to spatiotemporal inference suitable for use in AGI systems that incorporate logic-based components.

The subtlest and most difficult aspect of logical inference is not the logical rule-set nor the management of uncertainty, but the control of inference: the choice of which inference steps to take, in what order, in which contexts. Without effective inference control control methods, logical inference is an unscalable and infeasible approach to learning declarative knowledge. One of the key ideas underlying the CogPrime design is that inference control cannot effectively be handled by looking at logic alone. Instead, effective inference control must arise from the intersection between logical methods and other cognitive processes. In this chapter we describe some of the general principles used for inference control in the CogPrime design.

Having discussed inference in depth we now turn to other, simpler but equally important approaches to creating declarative knowledge.

CogPrime’s knowledge representation must be considered on two levels: implicit and explicit. This chapter considers mainly explicit knowledge representation, with a focus on representation of declarative knowledge.

One of the hallmarks of general intelligence is its capability to deal with novelty in its environment and/or goal-set. And dealing with novelty intrinsically requires creating novelty. It’s impossible to efficiently handle new situations without creating new ideas appropriately. Thus, in any environment complex and dynamic enough to support human-like general intelligence (or any other kind of highly powerful general intelligence), the creation of novel ideas will be paramount. New idea creation takes place in OpenCog via a variety of methods—e.g. inside MOSES which creates new program trees, PLN which creates new logical relationships, ECAN which creates new associative relationships, etc. But there is also a role for explicit, purposeful creation of new Atoms representing new concepts, outside the scope of these other learning mechanisms.

Among the many key features of the human brain omitted by typical formal neural network models, one of the foremost is the brain’s three-dimensionality. The brain is not just a network of neurons arranged as an abstract graph; it’s a network of neurons arranged in three-dimensional space, and making use of this three-dimensionality directly and indirectly in various ways and for various purposes.

This brief chapter deals with two important, coupled cognitive components of CogPrime: the component concerned with creating internal simulations of situations and episodes in the external physical world, and the one concerned with storing and retrieving memories of situations and episodes.

“Procedure learning”—learning step-by-step procedures for carrying out internal or external operations—is a highly critical aspect of general intelligence, and is carried out in CogPrime via a complex combination of methods.

In Chap. 2 we distinguished the explicit versus implicit aspects of knowledge representation in CogPrime. The explicit level consists of Atoms with clearly comprehensible meanings, whereas the implicit level consists of “maps”—collections of Atoms that become important in a coordinated manner, analogously to cell assemblies in an attractor neural net.

Language is a key aspect of human intelligence, and seems to be one of two critical factors separating humans from other intelligent animals—the other being the ability to use tools.

Two key approaches to endowing AGI systems with linguistic facility exist, as noted above.

The approach taken to NLP in the OpenCog project up through 2013, in practice, has involved engineering and integrating rule-based NLP systems as “scaffolding”, with a view toward later replacing the rule content with alternative content learned via an OpenCog system’s experience.

Language generation, unsurprisingly, shares most of the key features of language comprehension discussed in Chap. 26 —after all, the division between generation and comprehension is to some extent an artificial convention, and the two functions are intimately bound up both in the human mind and in the CogPrime architecture.

“Language” is an important abstraction—but one should never forget that it’s an abstraction. Language evolved in the context of embodied action, and even the most abstract language is full of words and phrases referring to embodied experience. Even our mathematics is heavily based on our embodied experience—geometry is about space; calculus is about space and time; algebra is a sort of linguistic manipulation generalized from experience-oriented language, etc.

We now turn to CogPrime’s representation and manipulation of procedural knowledge.

Language evolved for dialogue—not for reading, writing or speechifying. So it’s natural that dialogue is broadly considered a critical aspect of humanlike AGI—even to the extent that (for better or for worse) the conversational “Turing Test” is the standard test of human-level AGI.

By way of conclusion, we now return to the “key claims” that were listed at the end of Chap. 3 of Part 1. Quite simply, this is a list of claims such that—roughly speaking—if the reader accepts these claims, they should accept that the CogPrime approach to AGI is a viable one. On the other hand if the reader rejects one or more of these claims, they may well find one or more aspects of CogPrime unacceptable for some related reason. In Chap. 3 of Part. 1 we merely listed these claims; here we briefly discuss each one in the context of the intervening chapters, giving each one its own section or subsection.

This chapter begins the heart of the book: the part that explains how the CogPrime design aims to implement roughly human-like general intelligence, at the human level and ultimately beyond.

The critical factor shaping real-world general intelligence is resource constraint.

A significant portion of CogPrime’s dynamics is explicitly goal-driven—that is, based on trying (inasmuch as possible within the available resources) to figure out which actions will best help the system achieve its goals, given the current context.

Procedural knowledge must be learned, an often subtle and difficult process—but it must also be enacted. Procedure enaction is not as tricky a topic as procedure learning, but still is far from trivial, and involves the real-time interaction of procedures, during the course of execution, with other knowledge. In this brief chapter we explain how this process may be most naturally and flexibly carried out, in the context of CogPrime’s representation of procedures as programs (“Combo trees”).

Having discussed declarative, attentional, intentional and procedural knowledge, we are left only with sensorimotor and episodic knowledge to complete our treatment of the basic CogPrime “cognitive cycle” via which a CogPrime system can interact with an environment and seek to achieve its goals therein.

Many different approaches to “low-level” perception and action processing are possible within the overall CogPrime framework. We discussed several in the previous chapter, all elaborations of the general hierarchical pattern recognition approach.

This chapter focuses on the Neural Darwinism and component-system theory, and proposes a precise &quot;cognitive equation of motion&quot; governing the creative evolution of the network of neural processes. There is almost certainly no hope of methodically building a theory of mind from the neuron level up, based solely on empirical observations of neural behavior, without any 'leaps of level'. The brain is not predictable on the level of individual neuron behavior. The problem of pattern recognition, however, is too hard to be done by exhaustive search; some sort of insight is required to determine what sort of pattern to look for. In the Neural Darwinist view, the mind consists largely of a collection of 'neural maps', competing with one another and symbiotically reinforcing one another with all the complexity of an ecosystem. Edelman suggested that the smallest psychologically relevant entity is not the neuron but the neuronal group, the cluster of densely interconnected neurons containing say 100-100,000 neurons.

This chapter focuses on a nonlinear extension of Miller, Dollard, and Lewin's social conflict theory. It also focuses on the responsibility and capacity of the dynamical systems approach to contribute to the evolution of a balance between unity and diversity in psychology, to the exercise of free choice, and to the humanist missions of science, psychology, and society. Dynamics is properly a branch of mathematics. Sciences borrow the approach as a strategy to observe and model the behavior of complex sets of interrelated phenomena. Dynamical systems can change. The value of the control parameter may change by a very small amount and result in a radical transformation in the system's behavior. Such a major transformation is called a bifurcation. The mind of a social organization is also the dynamical system comprised of the interaction of the components of the social system with each other, its and their environments, and the personal minds of its participants.

A novel approach to the fully automated, unsupervised extraction of dependency grammars and associated syntax-to-semantic-relationship mappings from large text corpora is described. The suggested approach builds on the authors' prior work with the Link Grammar, RelEx and OpenCog systems, as well as on a number of prior papers and approaches from the statistical language learning literature. If successful, this approach would enable the mining of all the information needed to power a natural language comprehension and generation system, directly from a large, unannotated corpus.

CogPrime, a comprehensive architecture for embodied Artificial General Intelligence, is reviewed, covering the core architecture and algorithms, the underlying conceptual motivations, and the emergent structures, dynamics and functionalities expected to arise in a completely implemented CogPrime system once it has undergone appropriate experience and education. A qualitative argument is sketched, in favor of the assertion that a completed CogPrime system, given a modest amount of experience in an embodiment enabling it to experience a reasonably rich human-like world, will give rise to human-level general intelligence (with significant difference from humans, and with potential for progress beyond this level).

A novel “Mind-World Correspondence Principle” is proposed - which, given an environment and goal-set, heavily constrains the structure of any intelligent system capable of efficiently achieving those goals in that environment. This is proposed as a potential step toward a “general theory of general intelligence.” An approximate gloss of the proposed principle is: “For a mind to work intelligently toward certain goals in a certain world, there should be a nice mapping from goal-directed sequences of world-states into sequences of mind-states, where “nice” means that a world-state-sequence W composed of two parts W <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> and W <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> , gets mapped into a mind-state-sequence M composed of two corresponding parts M <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> and M <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> .” The principle is formulated using the mathematical language of category theory, but refinement of the principle into a precise theorem is left for later works. Discussion is given regarding the use of the principle to explain common properties of realworld intelligences such as the presence of hierarchical structure. Declarative, procedural and episodic memory systems, as present in human minds and human-like cognitive architectures, are formalized using category theory in a manner consistent with the Mind-World Correspondence principle. The notion of development of minds is similarly formulated, using the category-theoretic notion of natural transformations. It is suggested that this approach to cognitive analysis and modeling may eventually be useful for deriving and refining practical designs for Artificial General Intelligence.

Ray Kurzweil is well aware that the concept of a technological singularity (the idea of the emergence of a superintelligent technology that will take control) doesn't fit well into the average person's world view. So in his book, The Singularity Is Near, he presents an elaborate and careful argument in favor of his thesis that some sort of technological singularity is going to occur around the middle of this century. He's drawn a lot of exponential curves showing the rate of advance of various aspects of science and technology, reaching toward infinity in 2040-50. His Singularitarian vision is relentlessly positive, involving humans becoming integrated with various technologies - enhancing our brains and bodies, living as long as we want to, and fusing and networking with powerful artificially intelligent minds.

What does it mean for one mind to be a different version of another one, or a natural continuation of another one? Or put differently: when can two minds sensibly be considered versions of one another? This question occurs in relation to mind uploading, where one wants to be able to assess whether an approximate upload constitutes a genuine continuation of the uploaded mind or not. It also occurs in the context of the rapid mental growth that is likely to follow mind uploading, at least in some cases – here the question is, when is growth so rapid or discontinuous as to cause the new state of the mind to no longer be sensibly considerable as a continuation of the previous one?

We present the broad outlines of a roadmap toward human‐level artificial general intelligence (henceforth, AGI). We begin by discussing AGI in general, adopting a pragmatic goal for its attainment and a necessary foundation of characteristics and requirements. An initial capability landscape will be presented, drawing on major themes from developmental psychology and illuminated by mathematical, physiological, and information‐processing perspectives. The challenge of identifying appropriate tasks and environments for measuring AGI will be addressed, and seven scenarios will be presented as milestones suggesting a roadmap across the AGI landscape along with directions for future research and collaboration.

Intelligence (AI). The book series focuses on the establishment of new theories and paradigms in AI.At the same time, the series aims at exploring multiple scientific angles and methodologies, including results from research in cognitive science, neuroscience, theoretical and experimental AI, biology and from innovative interdisciplinary methodologies.

This chapter provides a general introduction to the volume, giving an overview of the AGI field and the current need for exploration and clarification of its foundations, and briefly summarizing the contents of the various chapters.

By exploring the relationships between different AGI architectures, one can work toward a holistic cognitive model of human-level intelligence. In this vein, here an integrative architecture diagram for human-like general intelligence is proposed, via merging of lightly modified version of prior diagrams including Aaron Sloman’s high-level cognitive model, Stan Franklin and the LIDA group’s model of working memory and the cognitive cycle, Joscha Bach and Dietrich Dorner’s Psi model of motivated action and cognition, James Albus’s three-hierarchy intelligent robotics model, and the author’s prior work on cognitive synergy in deliberative thought and metacognition, along with ideas from deep learning and computational linguistics.

In this paper, the OpenBiomind toolkit is used to apply GA, GP, and local search methods to analyze a large SNP dataset concerning Late-Onset Alzheimer’s Disease (LOAD). Classification models identifying LOAD with statistically significant accuracy are identified as well as using ensemble-based important features analysis in order to identify brain genes related to LOAD, most notably the solute carrier gene SLC6A15. Ensemble analysis is used to identify potentially significant interactions between genes in the context of LOAD.

Cognitive informatics is a transdisciplinary enquiry of computer science, information sciences, cognitive science, and intelligence science that investigates the internal information processing mechanisms and processes of the brain and natural intelligence, as well as their engineering applications in cognitive computing. Cognitive computing is an emerging paradigm of intelligent computing methodologies and systems based on cognitive informatics that implements computational intelligence by autonomous inferences and perceptions mimicking the mechanisms of the brain. This article presents a set of collective perspectives on cognitive informatics and cognitive computing, as well as their applications in abstract intelligence, computational intelligence, computational linguistics, knowledge representation, symbiotic computing, granular computing, semantic computing, machine learning, and social computing.

This is an interview I did with Ben Goertzel, an old friend and editor of the H+ electronic journal on technological, scientific, and cultural trends that are changing human beings. It's published by Humanity+, the world’s leading nonprofit for the ethical use of technology to extend human capabilities. It concerns some of my work in cybernetic technology and knowledge systems from the early 1990's, with some implications for my current work.

While it seems unlikely that any method of guaranteeing human-friendliness (“Friendliness”) on the part of advanced Artificial General Intelligence (AGI) systems will be possible, this doesn’t mean the only alternatives are throttling AGI development to safeguard humanity, or plunging recklessly into the complete unknown. Without denying the presence of a certain irreducible uncertainty in such matters, it is still sensible to explore ways of biasing the odds in a favorable way, such that newly created AI systems are significantly more likely than not to be Friendly. Several potential methods of effecting such biasing are explored here, with a particular but nonexclusive focus on those that are relevant to open-source AGI projects, and with illustrative examples drawn from the OpenCog open-source AGI project. Issues regarding the relative safety of open versus closed approaches to AGI are discussed and then nine techniques for biasing AGIs in favor of Friendliness are presented: 1. Engineer the capability to acquire integrated ethical knowledge. 2. Provide rich ethical interaction and instruction, respecting developmental stages. 3. Develop stable, hierarchical goal systems. 4. Ensure that the early stages of recursive self-improvement occur relatively slowly and with rich human involvement. 5. Tightly link AGI with the Global Brain. 6. Foster deep, consensus-building interactions between divergent viewpoints. 7. Create a mutually supportive community of AGIs. 8. Encourage measured co-advancement of AGI software and AGI ethics theory 9. Develop advanced AGI sooner not later. In conclusion, and related to the final point, we advise the serious co-evolution of functional AGI systems and AGI-related ethical theory as soon as possible, before we have so much technical infrastructure that parties relatively unconcerned with ethics are able to rush ahead with brute force approaches to AGI development.

A novel theory of reflective consciousness, will and self is presented, based on modeling each of these entities using self-referential mathematical structures called hypersets. Pattern theory is used to argue that these exotic mathematical structures may meaningfully be considered as parts of the minds of physical systems, even finite computational systems. The hyperset models presented are hypothesized to occur as patterns within the ”moving bubble of attention ” of the human brain and any brainlike AI system. They appear to be compatible with both panpsychist and materialist views of consciousness, and probably other views as well. 1

The hypothesis is presented that "cognitive synergy" -- proactive and mutually-assistive feedback between different cognitive processes associated with different types of memory -- may serve as a foundation for advanced artificial general intelligence. A specific AI architecture founded on this idea, OpenCogPrime, is described, in the context of its application to control virtual agents and robots. The manifestations of cognitive synergy in OpenCogPrime's procedural and declarative learning algorithms are discussed in some detail.

The general problem addressed in this book is a large and important one: how to usefully deal with huge storehouses of complex information about real-world situations. Every one of the major modes of interacting with such storehouses - querying, data mining, data analysis - is addressed by current technologies only in very limited and unsatisfactory ways. The impact of a solution to this problem would be huge and pervasive, as the domains of human pursuit to which such storehouses are acutely relevant is numerous and rapidly growing. Finally, we give a more detailed treatment of one potential

Having dealt with the representation of logical knowledge of various sorts, and briefly discussed the problem of translating nonlogical knowledge into logical knowledge, we now turn to the question of how large amounts of logical knowledge can pragmatically be stored. This chapter presents a brief and relatively nonmathematical interlude before we plunge into the more technical topics of mining patterns in logical knowledge stores, and carrying out inferences regarding changes and other patterns in logical knowledge stores.

Once one has stored a large knowledge base of logical relationships, then what? One can query the knowledge base – if one knows what one wants to ask for. One can carry out reasoning toward various goals. And another important question is how to find “unknown unknowns” – patterns in the knowledge base that are surprising and interesting yet unexpected. This quest goes by multiple names – data mining, pattern mining, information exploitation, and so forth. Whatever you call it, it’s a difficult challenge because in any large dataset, the number of possible patterns to search through is mind-boggling.

In this chapter we review the temporal and causal PLN relationship types and rules that are used to guide these sorts of inference in PLN, and give some simple examples to illustrate how they are used. In the following chapters we will present more elaborate examples of spatiotemporal reasoning, using these constructs and ideas.

In this chapter we consider the specific question of how the ideas of the previous chapters contribute to carrying out reasoning regarding the potential causes of salient changes in large knowledge stores.

As the discussion in the above chapters has hopefully made clear, every one of the tasks involved in logic-based RWR has been carefully approached by computer scientists, using various formalisms and software prototypes. We have produced more fully fleshed-out examples using PLN than other formalisms, because PLN is the inference framework we’re most familiar with and because we believe it most adequately integrates rich uncertainty representations with powerful inference mechanisms; but similar explorations could be produced using other inference formalisms, and of course one could also go into far greater detail than has been done above.

As emphasized from the start, the main purpose of this book has been to serve as a sort of “thought experiment.” We began with a very general problem of “querying, mining and analyzing huge real-world knowledge stores” – i.e., Real-World Reasoning.

Now we begin to dig into the nitty-gritty of our subject matter. Before discussing querying and analysis of complex, heterogeneous spatiotemporal and contextual knowledge, we must discuss representation of temporal knowledge (as well as, to a certain extent, spatial knowledge) . . . and before that, we must address knowledge representation in general.

Another major issue in formal logic systems is the quantification and incorporation of uncertainty. It is of course possible to represent uncertainty using theories within first-order logic, but many have argued that uncertainty is sufficiently basic to commonsense inference that it should be introduced into formal logic at a foundational level.

We’ve discussed the basics of logical reasoning and logical knowledge representation, in a general way. It’s now time to get more specific and talk about one of the applications of logic that’s most central to our topic: logical representation of time.

A large percentage of real-world knowledge pertains to space as well as time. Humans possess a great deal of evolved and learned common sense regarding representing and reasoning about space, but, this is not innate to software programs so we need to explicitly think about how to make our software systems space-friendly. From a logic point of view, space and time can in principle be treated just like any other sorts of relationships – but this turns out not to be an optimal point of view in terms of either human-friendliness or computational efficiency of logic systems. Rather, with space as with time, it is worthwhile to invest the effort to develop specialized techniques for handling spatial and spatiotemporal knowledge.

One of the critical issues confronted in doing logical inference on large spatiotemporal knowledge bases is the fact that most real-world logical relationships are contextual in nature. Contextuality must be explicitly taken into account to make real-world inference tractable. This is one of the issues with which formal-logic-based AI has traditionally had the most difficulty. In this chapter we briefly explore the notion of contextual inference and the key approaches that have been taken.

Many of the inferences one wants to draw about real-world, spatiotemporal, contextual knowledge involve cause and effect. But the relation between causation and logical inference is subtle and storied. As outlined above, deductive reasoning aims at deriving consequences (or effects, outcomes) from premises (or causes). Abductive reasoning aims at deriving possible causes from effects. Finally, inductive reasoning aims at deriving relationships between causes and effects, rules that lead from one to another. Causal reasoning is generally considered a form of inductive reasoning. More concretely, causal reasoning aims at an epistemological problem of establishing precise causal relationships between causes and effects, with focusing on detecting genuine, real causes for some effects, and genuine, real effects of some causes.

Logical methods are extremely powerful when supplied with appropriate data, but this begs the question of where the data comes from. How does data from the real world get into logical format in the first place?

A current project is briefly described, involving the application of the OpenCog integrative AGI system to control an animated agent in a videogame world inspired by the commercial game Minecraft. Among other purposes the project will provide a means of testing the hypothesis that cognitive synergy among learning mechanisms associated with different types of memory is valuable for embodied, human-like general intelligence.

Relational reasoning is a complex high-level cognitive function that should be part of a realistic computational equivalent of the human mind. People use relational reasoning often in everyday life in many different contexts (e.g., social understanding, political science, law, business). This paper discusses the application of relational reasoning to the recognition of geospatial patterns (e.g., clusters of buildings that constitute a facility). The relational reasoning model is based on cognitive science evidence and emerging neuroscience theory. Experiments show that the relational reasoning model can recognize geospatial patterns that have a significant degree of variation.

We describe a research project investigating “symbol grounding” – the dynamics by which psychological connections are made between abstract symbols and concrete physical phenomena observed via sense perception and motor action. The project involves the use of a 3D simulated environment (AGI-SIM) as a medium for training, teaching and developing an integrative, general-intelligence-oriented AI software system (the Novamente AI Engine). Via acting in AGI-SIM, Novamente forms concrete sensorimotor groundings for abstract relationships, such as those expressed in English by prepositions and subject-argument relationships. We describe results obtained using probabilistic-evolutionary learning within Novamente to learn groundings for the concept “near”, the use of these groundings in practical procedure learning (e.g. learning to play “fetch” in a simulated world), and we then discuss the correlation of these groundings with linguistic usage of related words, and the use of these groundings within analogical and other sorts of inference.

This chapter describes a 4 year research project (2008-2011) to build China’s first artificial brain. It takes an “evolutionary engineering” approach, by evolving 10,000s of neural net modules, (or “agents” in the sense of Minsky’s “Society of Mind” [Minsky 1988, 2007]), and connecting them to make artificial brains. These modules are evolved rapidly in seconds on a “Tesla” PC Supercomputer, and connected according to the artificial brain designs of human “BAs” (Brain Architects). The artificial brain will eventually contain thousands of pattern recognizer modules, and hundreds of decision modules that when suitably combined will control the hundreds of behaviors of a walking, talking robot.Request access from your librarian to read this chapter's full text.

Your fingers weave quick minarets... Speak in secret al-phabets... I light another cigarette... Learn to forget, learn to forget – Jim Morrison, ”Soul Kitchen” Of all the aspects differentiating lifelong learning from shorter-term, more specialized learning, perhaps none is more central than forgetting – or, to frame the issue more generally and technically, ”memory access speed deprioriti-zation. ” This extended abstract reviews some of the ideas in-volved in forgetting for lifelong learning systems, and briefly discusses the forgetting mechanisms used in the OpenCog integrative cognitive architecture. Defining Forgetting In ordinary human discourse, the word ”forget ” has multiple shades of meaning. It can refer to the irreversible elimination of a certain knowledge item

One approach to bridging the historic divide between ”symbolic” and ”subsymbolic” AI is to incorporate a subsymbolic system and a symbolic system into a synergetic integrative cognitive architecture. Here we consider various issues related to incorporating (subsymbolic) compositional spatiotemporal deep learning networks (CSDLNs, a term introduced to denote the category including HTM, DeSTIN and other similar systems) into an integrative cognitive architecture including symbolic aspects. The core conclusion is that for such integration to be meaningful, it must involve dynamic and adaptive linkage and conversion between CSDLN attractors spanning sensory, motor and goal hierarchies, and analogous representations in the remainder of the integrative architecture. We suggest the mechanism of ”semantic CSDLNs”, which maintain the general structure of CSDLNs but contain more abstract patterns, similar to those represented in more explicitly symbolic AI systems. This notion is made concrete by describing a planned integration of the DeSTIN CSDLN into the OpenCog integrative cognitive system (which includes a probabilistic-logical symbolic component).

This paper introduces a Partially Connected Neural Evolutionary (Parcone) model and its application to object and gesture recognition as part of the "China Brain Project", which aims to build China's first artificial brain, consisting of 10,000s of evolved interconnected neural net modules, each with its own little task, similar to Minsky's agents.

The Third Conference on Artificial General Intelligence was held March 5–8, 2010, at the University of Lugano in Lugano, Switzerland. The conference was chaired by Marcus Hutter and cochaired by Eric Baum and Emanuel Kitzelmann. The conference also provided a venue for the awarding of the 2010 Kurzweil Prizes.

A software architecture is described which enables a virtual agent in an online virtual world to carry out simple English language interactions grounded in its perceptions and actions. The use of perceptions to guide anaphor resolution is discussed, along with the use of natural language generation to answer simple questions about the observed world. This architecture has been implemented within the larger PetBrain system, which is built on the OpenCog open-source AI software framework and architected based on the OpenCogPrime design for integrative AGI, and has previously been used for nonlinguistic intelligent behaviors such as imitation and reinforcement learning.

Two new formal definitions of intelligence are presented, the "pragmatic general intelligence" and "efficient pragmatic general intelligence." Largely inspired by Legg and Hutter's formal definition of "universal intelligence," the goal of these definitions is to capture a notion of general intelligence that more closely models that possessed by humans and practical AI systems, which combine an element of universality with a certain degree of specialization to particular environments and goals. Pragmatic general intelligence measures the capability of an agent to achieve goals in environments, relative to prior distributions over goal and environment space. Efficient pragmatic general intelligences measures this same capability, but normalized by the amount of computational resources utilized in the course of the goal-achievement. A methodology is described for estimating these theoretical quantities based on observations of a real biological or artificial system operating in a real environment. Finally, a measure of the "degree of generality" of an intelligent system is presented, allowing a rigorous distinction between "general AI" and "narrow AI."

Spatiotemporal reasoning is an important skill that an AGI is expected to have, innately or not. Much work has already been done in defining reasoning systems for space, time and spacetime, such as the Region Connection Calculus for space, Allen's Interval Algebra for time, or the Qualitative Trajectory Calculus for motion. However, these reasoning systems rarely take adequate account of uncertainty, which poses an obstacle to using them in an AGI system confronted with an uncertain reality. In this paper we show how to use PLN (Probabilistic Logic Networks) to represent spatiotemporal knowledge and reasoning, via incorporating existing spatiotemporal calculi, and considering a novel extension of standard PLN truth values inspired by P(Z)-logic. This "PLN-ization" of existing spatiotemporal calculi, we suggest, constitutes an approach to spatiotemporal inference suitable for use in AGI systems that incorporate logic-based components.

Probabilistic Logic Networks (PLN), a comprehensive framework for uncertain inference currently in use in the OpenCog and Novamente Cognition Engine AGI software architectures, has previously been described in terms of the "experiential semantics" of an intelligent agent embodied in a world. However, several aspects of PLN are more easily interpreted and formulated in terms of "possible worlds semantics"; here we use a formal model of intelligent agents to show how a form of possible worlds semantics can be derived from experiential semantics, and use this to provide new interpretations of several aspects of PLN (including uncertain quantifiers, intensional inheritance, and indefinite probabilities.) These new interpretations have practical as well as conceptual benefits, as they give a unified way of specifying parameters that in the previous interpretations of PLN were viewed as unrelated.

This chapter describes a 4 year research project (2008-2011) to build China’s first artificial brain. It takes an “evolutionary engineering” approach, by evolving 10,000s of neural net modules, (or “agents” in the sense of Minsky’s “Society of Mind” [Minsky 1988, 2007]), and connecting them to make artificial brains. These modules are evolved rapidly in seconds on a “Tesla” PC Supercomputer, and connected according to the artificial brain designs of human “BAs” (Brain Architects). The artificial brain will eventually contain thousands of pattern recognizer modules, and hundreds of decision modules that when suitably combined will control the hundreds of behaviors of a walking, talking robot.

Cognitive informatics is a transdisciplinary enquiry of computer science, information sciences, cognitive science, and intelligence science that investigates the internal information processing mechanisms and processes of the brain and natural intelligence, as well as their engineering applications in cognitive computing. Cognitive computing is an emerging paradigm of intelligent computing methodologies and systems based on cognitive informatics that implements computational intelligence by autonomous inferences and perceptions mimicking the mechanisms of the brain. This article presents a set of collective perspectives on cognitive informatics and cognitive computing, as well as their applications in abstract intelligence, computational intelligence, computational linguistics, knowledge representation, symbiotic computing, granular computing, semantic computing, machine learning, and social computing.

In this paper, the OpenBiomind toolkit is used to apply GA, GP, and local search methods to analyze a large SNP dataset concerning Late-Onset Alzheimer’s Disease (LOAD). Classification models identifying LOAD with statistically significant accuracy are identified as well as using ensemble-based important features analysis in order to identify brain genes related to LOAD, most notably the solute carrier gene SLC6A15. Ensemble analysis is used to identify potentially significant interactions between genes in the context of LOAD.

Numerical optimization methods such as hillclimbing and simulated annealing have been applied to search for highly entangled multi-qubit states. Here the genetic algorithm is applied to this optimization problem -- to search not only for highly entangled states, but also for the corresponding quantum circuits creating these states. Simple quantum circuits for maximally (highly) entangled states are discovered for 3, 4, 5, and 6-qubit systems; and extension of the method to systems with more qubits is discussed. Among other results we have found explicit quantum circuits for maximally entangled 5 and 6-qubit circuits, with only 8 and 13 quantum gates respectively. One significant advantage of our method over previous ones is that it allows very simple construction of quantum circuits based on the quantum states found.

A number of leading cognitive architectures that are inspired by the human brain, at various levels of granularity, are reviewed and compared, with special attention paid to the way their internal structures and dynamics map onto neural processes. Four categories of Biologically Inspired Cognitive Architectures (BICAs) are considered, with multiple examples of each category briefly reviewed, and selected examples discussed in more depth: primarily symbolic architectures (e.g. ACT-R), emergentist architectures (e.g. DeSTIN), developmental robotics architectures (e.g. IM-CLEVER), and our central focus, hybrid architectures (e.g. LIDA, CLARION, 4D/RCS, DUAL, MicroPsi, and OpenCog). Given the state of the art in BICA, it is not yet possible to tell whether emulating the brain on the architectural level is going to be enough to allow rough emulation of brain function; and given the state of the art in neuroscience, it is not yet possible to connect BICAs with large-scale brain simulations in a thoroughgoing way. However, it is nonetheless possible to draw reasonably close function connections between various components of various BICAs and various brain regions and dynamics, and as both BICAs and brain simulations mature, these connections should become richer and may extend further into the domain of internal dynamics as well as overall behavior.

One of the major differences between irregular warfare (IW) simulation and conventional warfare simulation is the differing data requirements. The data required by conventional warfare simulation is very narrow and well defined compared to that required by IW simulation. In fact, almost any data that is remotely related to social conditions on the Internet and beyond may be used in IW simulation. There is no problem with the existence of data for simulations in this information age. The problem is that this data is not expressed in a way that can be used for simulations. This paper is about a natural language processing program, Indra, which works towards solving this problem by making data available for multiple purposes other than those for which it was first intended.

We outline a general conceptual definition of real-world general intelligence that avoids the twin pitfalls of excessive mathematical generality, and excessive anthropomorphism. Drawing on prior literature, a definition of general intelligence is given, which defines the latter by reference to an assumed measure of the simplicity of goals and environments. The novel contribution presented is to gauge the simplicity of an entity in terms of the ease of communicating it within a community of embodied agents (the so-called embodied communication prior or ECP). Augmented by some further assumptions about the statistical structure of communicated knowledge, this choice is seen to lead to a model of intelligence in terms of distinct but interacting memory and cognitive subsystems dealing with procedural, declarative, sensory/episodic, attentional and intentional knowledge. A sister paper then extends these ideas to yield a "cognitive synergy theory" that suggests specific conclusions for the architecture of artificial general intelligences, based on the ECP.

Do there exist general principles, which any system must obey in order to achieve advanced general intelligence using feasible computational resources? Here we propose one candidate: ldquocognitive synergy,rdquo a principle which suggests that general intelligences must contain different knowledge creation mechanisms corresponding to different sorts of memory (declarative, procedural, sensory/episodic, attentional, intentional); and that these different mechanisms must be interconnected in such a way as to aid each other in overcoming memory type specific combinatorial explosions.

The OpenBiomind toolkit is used to apply GA, GP and local search methods to analyze a large SNP dataset concerning late-onset Alzheimer's disease (LOAD). Classification models identifying LOAD with statistically significant accuracy are identified, and ensemble-based important features analysis is used to identify brain genes related to LOAD, most notably the solute carrier gene SLC6A15. Ensemble analysis is used to identify potentially significant interactions between genes in the context of LOAD.

OpenCogPrime (OCP), a comprehensive architecture for artificial general intelligence (AGI) is briefly overviewed. Aimed in the long term at AGI at the human level and beyond, the current partial implementation of OCP is being used for applications such as controlling virtual pets in virtual worlds and inferring novel conclusions from sets of semantic relations extracted from natural language. The key aspects of OCP are described here in the context of the theoretical foundation of "cognitive synergy theory"; and the current implementation status is briefly reviewed.

Cognitive Informatics (CI) is a transdisciplinary enquiry of computer science, information sciences, cognitive science, and intelligence science that investigates into the internal information processing mechanisms and processes of the brain and natural intelligence, as well as their engineering applications in cognitive computing. This paper summarizes the presentations of a set of 14 position papers in the ICCI'09 Panel on Cognitive Informatics and Cognitive Computing contributed from invited panelists who are part of the world's renowned researchers and scholars in the field of cognitive informatics and cognitive computing.

The First Conference on Artificial General Intelligence (AGI‐08) was held on March 1–3, 2008, at the University of Memphis. The overall goal of the conference was to work toward a common understanding of the most promising paths toward creating AI systems with general intelligence at the human level and beyond and to share interim results and ideas achieved by researchers actively working toward powerful artificial general intelligence.

This article reports on AGI‐09, the Second International Conference on Artificial General Intelligence, which was held 6‐9 March 2009 in Arlington, Virginia, USA. The purpose of the conference was to exchange ideas about the creation of artificial systems with general intelligence at, and ultimately beyond, the human level.

The First Conference on Artificial General Intelligence (AGI-08) was held on March 1-3, 2008, at the University of Memphis. The overall goal of the conference was to work toward a common understanding of the most promising paths toward creating AI systems with general intelligence at the human level and beyond, and to share interim results and ideas achieved by researchers actively working toward powerful artificial general intelligence.

A novel method for simultaneously storing memories and allocating resources in AI systems is presented. The method, Economic Attention Networks (ECANs), bears some resemblance to the spread of activation in attractor neural networks, but differs via explicitly differentiating two kinds of "activation" (Short Term Importance, related to processor allocation; and Long Term Importance, related to memory allocation), and in using equations that are based on ideas from economics rather than approximative neural modeling. Here we explain the basic ideas of ECANs, and then investigate the functionality of ECANs as associative memories, via mathematical analysis and the reportage of experimental results obtained from the implementation of ECANs in the OpenCog integrative AGI system.

A class of environments for teaching and evaluating AGI systems is described, modeled loosely on preschools used for teaching human children and intended specifically for early-stage systems aimed at approaching human-level, human-inspired AGI through a process resembling human developmental psychology. Some particulars regarding implementation of environments in this class in online virtual worlds are presented, along with details on associated evaluation tasks, and discussion of environments potentially appropriate for "AGI preschool graduates," based on more advanced human schooling.

Traditional machine learning systems work with relatively flat, uniform data representations, such as feature vectors, time-series, and context-free grammars. However, reality often presents us with data which are best understood in terms of relations, types, hierarchies, and complex functional forms. One possible representational scheme for coping with this sort of complexity is computer programs. This immediately raises the question of how programs are to be best represented. We propose an answer in the context of ongoing work towards artificial general intelligence.

The "China Brain Project" is a 4 year (2008)(2009)(2010)(2011), 10.5 million RMB research project to build China's first artificial brain, which will consist of 10,000-50,000 neural net modules which are evolved rapidly in special FPGA hardware, downloaded one by one into a PC or supercomputer, and then connected according to the designs of human "BAs" (brain architects) to build an artificial brain with thousands of pattern detectors to control the hundreds of behaviors of a two legged robot.

Artificial General Intelligence (AGI) research focuses on the original and ultimate goal of AI – to create broad human-like and transhuman intelligence, by exploring all available paths, including theoretical and experimental computer science, cognitive science, neuroscience, and innovative interdisciplinary methodologies. Due to the difficulty of this task, for the last few decades the majority of AI researchers have focused on what has been called narrow AI – the production of AI systems displaying intelligence regarding specific, highly constrained tasks. In recent years, however, more and more researchers have recognized the necessity – and feasibility – of returning to the original goals of the field. Increasingly, there is a call for a transition back to confronting the more difficult issues of human level intelligence and more broadly artificial general intelligence. The Conference on Artificial General Intelligence is the only major conference series devoted wholly and specifically to the creation of AI systems possessing general intelligence at the human level and ultimately beyond. Its second installation, AGI-09, in Arlington, Virginia, March 6-9, 2009, attracted 67 paper submissions, which is a substantial increase from the previous year. Of these submissions, 33 (i.e., 49%) were accepted as full papers for presentation at the conference. Additional 13 papers were included as position papers. The program also included a keynote address by J¨urgen Schmidhuber on The New AI, a post-conference workshop on The Future of AI, and a number of pre-conference tutorials on various topics related to AGI.

Among those who believe that richly embodied AGI is a promising path to creating AGI systems displaying human-level general intelligence, the possibility of virtual-world embodiment, as opposed to real-world robotic embodiment holds considerable appeal. Here we consider the question of what properties a virtual world should have in order to constitute an adequate environment for the cognitive development of a human-like, human-level general intelligence. We ask what properties a virtual world must have so that an AGI embodied in that world could viably infer humanlike theories of naive physics and folk psychology, and carry out tasks typically required in cognitive development tasks and preschool play centers. Based on these considerations we suggest a “minimal adequate environment” we call “BlocksNBeadsWorld,” in which agents can construct objects from blocks using adhesives, and can also fill containers, coat objects and create fabrics and substances with various sorts of differentially adhesive beads.

A deeply-interactive hybrid neural-symbolic cognitive architecture is defined as one in which the neural-net and symbolic components interact frequently and dynamically, so that each intervenes significantly in the other&amp;apos;s internal operations, and the two form a combined dynamical system at the time-scale of each component&amp;apos;s individual cognitive operations. An example architecture of this nature that is currently under development is described: OpenCog NS, based on integration of the OpenCog cognitive architecture (which incorporates symbolic, evolutionary and connectionist aspects) with a hierarchical attractor neural network (HANN). In this integrated architecture, the neural and non-neural aspects each play major roles, and the depth of the interconnection is revealed for example in the facts that symbolic reasoning intervenes in the process of attractor formation within the HANN, whereas the HANN plays a major role in guiding the individual steps of logical inference and evolutionary program learning processes.

Novel artificial intelligence methodologies were applied to analyze gene expression microarray data gathered from mice under a calorie restriction (CR) regimen. The data were gathered from three previously published mouse studies; these datasets were merged together into a single composite dataset for the purpose of conducting a broader-based analysis. The result was a list of genes that are important for the impact of CR on lifespan, not necessarily in terms of their individual actions but in terms of their interactions with other genes. Furthermore, a map of gene interrelationships was provided, suggesting which intergene interactions are most important for the effect of CR on life extension. In particular our analysis showed that the genes Mrpl12, Uqcrh, and Snip1 play central roles regarding the effects of CR on life extension, interacting with many other genes (which the analysis enumerates) in carrying out their roles. This is the first time that the genes Snip1 and Mrpl12 have been identified in the context of aging. In a follow-up analysis aimed at validating these results, the analytic process was rerun with a fourth dataset included, yielding largely comparable results. Broadly, the biological interpretation of these analytical results suggests that the effects of CR on life extension are due to multiple factors, including factors identified in prior theories of aging, such as the hormesis, development, cellular, and free radical theories.

Current work is described wherein simplified versions of the Novamente cognition engine (NCE) are being used to control virtual agents in virtual worlds such as game engines and second life. In this context, an IRC (imitation-reinforcement-correction) methodology is being used to teach the agents various behaviors, including simple tricks and communicative acts. Here we describe how this work may potentially be exploited and extended to yield a pathway toward giving the NCE robust, ultimately human level natural language conversation capability. The pathway starts via using the current system to instruct NCE-controlled agents in semiosis and gestural communication; and then continues via integration of a particular sort of hybrid rule-based/statistical NLP system (which is currently partially complete) into the NCE-based virtual agent system, in such a way as to allow experiential adaptation of the rules underlying the NLP system, in a manner that builds on the agentpsilas knowledge of semiosis and gesture.

A teaching methodology called Imitative-Reinforcement-Corrective (IRC) learning is described, and proposed as a general approach for teaching embodied non-linguistic AGI systems. IRC may be used with a variety of different learning algorithms, but it is particularly easily described in EC lingo. In these terms, it is a framework for automatically learning a procedure that generates a desired type of behavior, in which: a set of exemplars of the target behavior-type are utilized for fitness estimation; reinforcement signals from a human teacher are used for fitness evaluation; and the execution of candidate procedures may be modified by the teacher via corrections delivered in real-time. An example application of IRC to teach behaviors to AI-controlled artificial animals embodied in the Second Life virtual world is described in detail, including a review of the overall virtual-animal-control software architecture and how the integrative teaching/learning methodology fits into it. In this example application architecture, the learning algorithm may be toggled between hillclimbing and probabilistic evolutionary learning. Envisioned future applications are also discussed, including an application to embodied language learning applicable to agents in Second Life and other virtual worlds.

Abstract: A series of hypotheses is proposed, connecting neural structures and dynamics with the formal structures and processes of probabilistic logic. First, a

Abstract. A novel theory of the stages of ethical development in intelligent systems is proposed, incorporating prior related theories by Kohlberg and Gilligan, as well as Piaget’s theory of cognitive development. This theory is then applied to the ethical development of integrative AGI systems that contain components carrying out simulation and uncertain inference – the key hypothesis being that effective integration of these components is central to the ascent of the AGI system up the ethical-stage hierarchy.

Indefinite probabilities are a novel technique for quantifying uncertainty, which were created as part of the PLN (Probabilistic Logic Networks) logical inference engine, which is a key component of the Novamente Cognition Engine (NCE), an integrative AGI system. Previous papers have discussed the use of indefinite probabilities in the context of a variety of logical inference rules, but have omitted discussion of quantification. Here this gap is filled, and a mathematical procedure is provided allowing the propagation of indefinite probabilities through universal and existential quantifiers, and also through a variety of fuzzy quantifiers corresponding to natural language quantifiers (such as “few”, “many”, “a lot”, “hardly any”, etc.). Proper probabilistic handling of various quantifier transformation rules is also discussed. Together with the ideas in prior publications, and a forthcoming sequel paper on indefinite probabilities for intensional inference, these results allow probabilistic logic based on indefinite probabilities to be utilized for the full scope of inferences involved in intelligent reasoning. To illustrate the ideas and algorithms involved, we give two concrete examples: Halpern's “crooked lottery” predicate, and a commonsense syllogism that uses fuzzy quantifiers together with the standard PLN term logic deduction rule.

Abstract. Recent psychological research suggests that the individual human mind may be effectively modeled as involving a group of interacting social actors: both various subselves representing coherent aspects of personality; and virtual actors embodying “internalizations of others. ” Recent neuroscience research suggests the further hypothesis that these internal actors may in many cases be neurologically associated with collections of mirror neurons. Taking up this theme, we study the mathematical and conceptual structure of sets of inter-observing actors, noting that this structure is mathematically isomorphic to the structure of physical entities called “mirrorhouses.” Mirrorhouses are naturally modeled in terms of abstract algebras such as quaternions and octonions (which also play a central role in physics), which leads to the conclusion that the presence within a single human mind of multiple inter-observing actors naturally gives rise to a mirrorhouse-type cognitive structure and hence to a quaternionic and octonionic algebraic structure as a significant aspect of human intelligence. Similar conclusions would apply to nonhuman intelligences such as AI’s, we suggest, so long as these intelligences included empathic social modeling (and/or other cognitive dynamics leading to the creation of simultaneously active subselves or other internal autonomous actors) as a significant component.

The XIA-MAN architecture for intelligent humanoid robot control is proposed, a novel design in which perception and action are achieved via a combination of GA-evolved neural-net modules with existing open-source software packages; and cognition is achieved via the OpenCog Prime framework. XML is used to communicate between components, enabling simple pluggability of additional or modified components, and leading to the name XIA-MAN (eXtensible Integrative Artificial Man). XIA-MAN’s neural net modules are used to convert between high-dimensional numerical representations denoting perceptions and actions, and probabilistic symbolic representations suitable for cognitive manipulation. XIA-MAN’s Cognition Engine is used, at each time cycle, to choose a high-level behavioral procedure that is believes is likely to enable the robot to achieve its goals in the current context. This provides a pragmatic approach to achieving intelligent robot functionality given currently available technologies; and the architecture is conceptually reminiscent of the complexly interconnected multimodular architecture of the brain. Initial work involves an incarnation of the XIA-MAN architecture using the Nao humanoid robot, to create an intelligent humanoid called XiaoNao.

The field of Artificial Intelligence (AI) was initially directly aimed at the construction of 'thinking machines' - that is, computer systems with human-like general intelligence. But this task proved more difficult than expected. As the years passed, AI researchers gradually shifted focus to producing AI systems that intelligently approached specific tasks in relatively narrow domains. In recent years, however, more and more AI researchers have recognized the necessity - and the feasibility - of returning to the original goal of the field. Increasingly, there is a call to focus less on highly specialized 'narrow AI' problem solving systems, and more on confronting the difficult issues involved in creating 'human-level intelligence', and ultimately general intelligence that goes beyond the human level in various ways. Artificial General Intelligence (AGI), as this renewed focus has come to be called, attempts to study and reproduce intelligence as a whole in a domain independent way. Encouraged by the recent success of several smaller-scale AGI-related meetings and special tracks at conferences, the initiative to organize the very first international conference on AGI was taken, with the goal to give researchers in the field an opportunity to present relevant research results and to exchange ideas on topics of common interest. In this collection you will find the conference papers: full-length papers, short position statements and also the papers presented in the post conference workshop on the sociocultural, ethical and futurological implications of AGI.IOS Press is an international science, technical and medical publisher of high-quality books for academics, scientists, and professionals in all fields. Some of the areas we publish in: -Biomedicine -Oncology -Artificial intelligence -Databases and information systems -Maritime engineering -Nanotechnology -Geoengineering -All aspects of physics -E-governance -E-commerce -The knowledge economy -Urban studies -Arms control -Understanding and responding to terrorism -Medical informatics -Computer Sciences

The OpenCog software development framework, for advancement of the development and testing of powerful and responsible integrative AGI, is described. The OpenCog Framework (OCF) 1.0, to be released in 2008 under the GPLv2, is comprised of a collection of portable libraries for OpenCog applications, plus an initial collection of cognitive algorithms that operate within the OpenCog framework. The OCF libraries include a flexible knowledge representation embodied in a scalable knowledge store, a cognitive process scheduler, and a plug-in architecture for allowing interaction between cognitive, perceptual, and control algorithms.

A novel approach, model-based clustering, is described foridentifying complex interactions between genes or gene-categories based on static gene expression data. The approach deals with categorical data, which consists of a set of gene expressionprofiles belonging to one category, and a set belonging to anothercategory. An evolutionary algorithm (Meta-Optimizing Semantic Evolutionary Search, or MOSES) is used to learn an ensemble of classification models distinguishing the two categories, based on inputs that are features corresponding to gene expression values. Each feature is associated with a model-based vector, which encodes quantitative information regarding the utilization of the feature across the ensembles of models. Two different ways of constructing these vectors are explored. These model-based vectors are then clustered using a variant of hierarchical clustering called Omniclust. The result is a set of model-based clusters, in which features are gathered together if they are often considered together by classification models -- which may be because they're co-expressed, or may be for subtler reasons involving multi-gene interactions. The method is illustrated by applying it to two datasets regarding human gene expression, one drawn from brain cells and pertinent to the neurogenetics of aging, and the other drawn from blood cells and relating to differentiating between types of lymphoma. We find that, compared to traditional expression-based clustering, the new method often yields clusters that have higher mathematical quality (in the sense of homogeneity and separation) and also yield novel and meaningful insights into the underlying biological processes.

No abstract available.

We describe a research project investigating “symbol grounding” – the dynamics by which psychological connections are made between abstract symbols and concrete physical phenomena observed via sense perception and motor action. The project involves the use of a 3D simulated environment (AGI-SIM) as a medium for training, teaching and developing an integrative, general-intelligence-oriented AI software system (the Novamente AI Engine). Via acting in AGI-SIM, Novamente forms concrete sensorimotor groundings for abstract relationships, such as those expressed in English by prepositions and subject-argument relationships. We describe results obtained using probabilistic-evolutionary learning within Novamente to learn groundings for the concept “near”, the use of these groundings in practical procedure learning (e.g. learning to play “fetch” in a simulated world), and we then discuss the correlation of these groundings with linguistic usage of related words, and the use of these groundings within analogical and other sorts of inference.

A novel theory of stages in cognitive development is presented, loosely corresponding to Piagetan theory but specifically oriented toward AI systems centered on uncertain inference components. Four stages are articulated (infantile, concrete, formal and reflexive), and are characterized both in terms of external cognitive achievements (a la Piaget) and in terms of internal inference control dynamics. The theory is illustrated via the analysis of specific problem solving tasks corresponding to the different stages. The Novamente AI Engine, with its Probabilistic Logic Networks uncertain inference component and its embodiment in the AGI-SIM simulation world, is used as an example throughout. I.

The Novamente Cognition Engine (NCE) architecture for Artificial General Intelligence is briefly reviewed, with a focus on exploring how the various cognitive processes involved in the architecture are intended to cooperate in carrying out moderately complex tasks involving controlling an agent embodied in the AGI-Sim 3D simulation world. A handful of previous conference papers have reviewed the overall architecture of the NCE, and discussed some accomplishments of the current, as yet incomplete version of the system; this paper is more speculative and focuses on the intended behaviors of the NCE once the implementation of all its major cognitive processes is complete. The “iterated Easter Egg Hunt” scenario is introduced and used as a running example throughout, due to its combination of perceptual, physical-action, social and self-modeling aspects. To aid in explaining the intended behavior of the NCE, a systematic typology of NCE cognitive processes is introduced. Cognitive processes are typologized as global, operational or focused; and, the focused processes are more specifically categorized as either forward-synthesis or backward-synthesis processes. The typical dynamics of focused cognition is then modeled as an ongoing oscillation between forward and backward synthesis processes, with critical emergent structures such as self and consciousness arising as attractors of this oscillatory dynamic. The emergence of models of self and others from this oscillatory dynamic is reviewed, along with other aspects of cognitive-process integration in the NCE, in the context of the iterated Easter Egg Hunt scenario.

We show here an example of the application of a novel method, MUTIC (model utilization-based clustering), used for identifying complex interactions between genes or gene categories based on gene expression data. The method deals with binary categorical data which consist of a set of gene expression profiles divided into two biologically meaningful categories. It does not require data from multiple time points. Gene expression profiles are represented by feature vectors whose component features are either gene expression values, or averaged expression values corresponding to gene ontology or protein information resource categories. A supervised learning algorithm (genetic programming) is used to learn an ensemble of classification models distinguishing the two categories based on the feature vectors corresponding to their members. Each feature is associated with a "model utilization vector", which has an entry for each high-quality classification model found, indicating whether or not the feature was used in that model. These utilization vectors are then clustered using a variant of hierarchical clustering called Omniclust. The result is a set of model utilization-based clusters, in which features are gathered together if they are often considered together by classification models - which may be because they are co-expressed, or may be for subtler reasons involving multi-gene interactions. The MUTIC method is illustrated here by applying it to a dataset regarding gene expression in prostate cancer and control samples. Compared to traditional expression-based clustering, MUTIC yields clusters that have higher mathematical quality (in the sense of homogeneity and separation) and that also yield novel insights into the underlying biological processes.

This introductory chapter sets the stage for the book as a whole. First, the purpose of the conference that led to the book is reviewed. Then, the notion of “Artificial General Intelligence” (AGI) is clarified, including a brief survey of the past and present situation of the field, an analysis and refutation of some common objections and doubts regarding the AGI area of research, and a discussion of what needs to be addressed by the field as a whole in the near future. Finally, there is a summary of the contents of the other chapters in the book.

domains such as theorem-proving and linguistic semantics. In the Novamente AGI architecture, however, probabilistic logic is used for a wider variety of purposes, including simple reinforcement learning of infantile behaviors, which are primarily concerned with perception and action rather than abstract cognition. This paper reports some simple experiments designed to validate the viability of this approach, via using the PLN probabilistic logic framework, implemented within the Novamente AGI architecture, to carry out reinforcement learning of simple embodied behaviors in a 3D simulation world (AGISim). The specific experiment focused upon involves teaching Novamente to play the game of “fetch ” using reinforcement learning based on repeated partial rewards. Novamente is an integrative AGI architecture involving considerably more than just PLN; however, in this “fetch ” experiment, the only cognitive process PLN is coupled with is simple perceptual pattern mining; other Novamente cognitive processes such as evolutionary learning and economic attention allocation are not utilized, so as to

During his talk at the 2006 AGI Workshop, Stan Franklin suggested that his LIDA architecture might fruitfully be considered not only as a specific AGI design, but also as a general framework within which to discuss and compare various AGI designs and approaches. With this in mind, following the workshop itself, Ben Goertzel formulated a list of simple questions intended to be pertinent to any AGI software design, mostly based on the conceptual framework presented in Stan Franklin's workshop presentation, with a couple additions and variations. Four individuals (Goertzel, Franklin, Pei Wang and Alexei Samsonovich) took up this offer, and their answers are reported here, without modification. The result is an interesting and novel perspective of a significant slice of contemporary AGI research.

At the end of the 2006 AGI Workshop, a number of presenters were invited to participate in a panel discussion on the theme “What Are the Bottlenecks, and How Soon to AGI?” This chapter presents a record of that dialogue, including audience participation, lightly edited by the presenters themselves for greater readability.

Analysis of postgenomic biological data (such as microarray and SNP data) is a subtle art and science, and the statistical methods most commonly utilized sometimes prove inadequate. Machine learning techniques can provide superior understanding in many cases, but are rarely used due to their relative complexity and obscurity. A challenge, then, is to make machine learning approaches to data analysis available to the average biologist in a user-friendly way. This challenge is addressed by the Biomind ArrayGenius product, an easy-to-use Web-based system providing microarray analysis based on genetic programming, kernel methods, and incorporation of knowledge from biological ontologies; and GeneGenius, its sister product for SNP data. This paper focuses on the obstacles faced and lessons learned in the course of creating, deploying, maintaining and selling ArrayGenius and GeneGenius – many of which are generic to any effort involving the creation of complex AI-based products addressing complex domain problems.

It is well known that the Clifford Algebras, and their quaternionic and octonionic subalgebras, are structures of fundamental importance in modern physics. Geoffrey Dixon has even used them as the centerpiece of a novel approach to Grand Unification. In the spirit of Wheeler's notion of pregeometry and more recent work on quantum set theory, the goal of the present investigation is to explore how these algebras may be seen to emerge from a simpler and more primitive order. In order to observe this emergence in the most natural way, a pregeometric domain is proposed that consists of two different kinds of boundaries, each imposing different properties on the combinatory operations occurring between elements they contain. It is shown that a very simple variant of this kind of multiboundary gives rise to Clifford Algebra, in much the same way as Spencer-Brown's simpler single-boundary algebra gives rise to Boolean algebra.

The creation of robust mechanisms for uncertain inference is central to the development of Artificial General Intelligence systems. While probability theory provides a principled foundation for uncertain inference, the mathematics of probability theory has not yet been developed to the point where it is possible to handle every aspect of the uncertain inference process in practical situations using rigorous probabilistic calculations. Due to the need to operate within realistic computational resources, probability theory presently requires augmentation with heuristics in order to be pragmatic for general intelligence (as well as for other purposes such as large-scale data analysis). The authors have been involved with the creation of a novel, general framework for pragmatic probabilistic inference in an AGI context, called Probabilistic Logic Networks (PLN). PLN integrates probability theory with a variety of heuristic inference mechanisms; it encompasses a rich set of first-order and higher-order inference rules, and it is highly flexible and adaptive, and easily configurable. This paper describes a single, critical aspect of the PLN framework, which has to with the quantification of uncertainty. In short, it addresses the question: What should an uncertain truth value be, so that a general intelligence may use it for pragmatic reasoning? We propose a new approach to quantifying uncertainty via a hybridization of Walley's theory of imprecise probabilities and Bayesian credible intervals. This “indefinite probability” approach provides a general method for calculating the “weight-of-evidence” underlying the conclusions of uncertain inferences. Moreover, both Walley's imprecise beta-binomial model and standard Bayesian inference can be viewed mathematically as special cases of the more general indefinite probability model. Via exemplifying the use of indefinite probabilities in a variety of PLN inference rules (including exact and heuristic ones), we argue that this mode of quantifying uncertainty may be adequate to serve as an ingredient of powerful artificial general intelligence.

A novel method, MUTIC (model utilization-based clustering), is described for identifying complex interactions between genes or gene-categories based on gene expression data. The method deals with binary categorical data, which consists of a set of gene expression profiles divided into two biologically meaningful categories. It does not require data from multiple time points. Gene expression profiles are represented by feature vectors whose component features are either gene expression values, or averaged expression values corresponding to gene ontology or protein information resource categories. A supervised learning algorithm (genetic programming) is used to learn an ensemble of classification models distinguishing the two categories based on the feature vectors corresponding to their members. Each feature is associated with a "model utilization vector," which has an entry for each high-quality classification model found, indicating whether or not the feature was used in that model. These utilization vectors are then clustered using a variant of hierarchical clustering called Omniclust. The result is a set of model-utilization-based clusters, in which features are gathered together if they are often considered together by classification models - which may be because they're co-expressed, or may be for subtler reasons involving multi-gene interactions. The MUTIC method is illustrated via applying it to a dataset regarding gene expression in human brains of various ages. Compared to traditional expression-based clustering, MUTIC yields clusters that have higher mathematical quality (in the sense of homogeneity and separation) and also yield novel insights into the underlying biological processes.

It is proposed that the creation of artificial general intelligence (AGI) at the human level and ultimately beyond is a problem addressable via integrating computer science algorithms and data structures within a cognitive architecture oriented toward experiential learning. A general conceptual framework for AGI is presented, beginning with a philosophy of mind based on the concept of pattern, then moving to a general mathematical and conceptual framework for modeling intelligent systems, self-modifying evolving probabilistic hypergraphs (SMEPH), and finally to an overview of a specific design for AGI, the Novamente AI engine. The problem of teaching an AGI system is discussed, in the context of Novamente's embodiment in the AGI-SIM simulation world. An educational program based loosely on Piaget's developmental stages is outlined, followed by more detailed consideration of the learning by Novamente in AGI-SIM of the Piagetan infant-level capability of "object permanence".

A novel technique is described, wherein Support Vector Machines are used to perform relatively effective text categorization based on small numbers of positive examples (fewer than 10 in some cases). It is assumed that in addition to the positive examples a query describing the positive category is given (in the form of a set of key phrases or a sentence). The technique combines two innovations: a special way of altering the SVM score threshold based on looking at the distribution of scores across the training set; and, a method of feature selection that involves retaining only features that display semantic association to the content words in the query (according to a word-association database produced by statistical analysis of a parsed corpus). Examples are given on a number of test cases drawn from the Reuters and FBIS news archives.

Support Vector Machines are used to combine the outputs of multiple entity extractors, thus creating a composite entity extraction system. The composite system has a significantly higher f-measure than any of the component systems. Compared to a standard voting technique for combining the results of multiple entity extractors, the SVM approach produces comparable precision and recall statistics but tends to utilize fewer of the component entity extractors, thus providing superior computational efficiency, which is critical in practical applications. In this paper, we present our experimental results of comparing a standard voting technique with SVM that each aggregate four entity extractors. We also describe our future plans of integrating agent-based technology into our experimental testbed where we examine the evolution of composite techniques as part of the analysis stream. Given that much of the improvement comes from tuning the algorithms to the data stream with a human-in-the-loop, we are considering the merits of employing cognitive agents that are strategically embedded in the workflow for processing data. As we tune the algorithms for better performance on the data streams, we envision agents learning the patterns of data streams and apply the appropriate tuning to ensure optimality.

We describe BioLiterate, a prototype software system which infers relationships involving relationships between genes, proteins and malignancies from research abstracts, and has initially been tested in the domain of the molecular genetics of oncology. The architecture uses a natural language processing module to extract entities, dependencies and simple semantic relationships from texts, and then feeds these features into a probabilistic reasoning module which combines the semantic relationships extracted by the NLP module to form new semantic relationships. One application of this system is the discovery of relationships that are not contained in any individual abstract but are implicit in the combined knowledge contained in two or more abstracts.

We describe BioLiterate, a prototype software system which infers relationships involving relationships between genes, proteins and malignancies from research abstracts, and has initially been tested in the domain of the molecular genetics of oncology. The architecture uses a natural language processing module to extract entities, dependencies and simple semantic relationships from texts, and then feeds these features into a probabilistic reasoning module which combines the semantic relationships extracted by the NLP module to form new semantic relationships. One application of this system is the discovery of relationships that are not contained in any individual abstract but are implicit in the combined knowledge contained in two or more abstracts.

The Gene Ontology (GO) database annotates a large number of genes according to their functions (the biological processes, molecular functions and cellular components in which they are involved). However, it is far from complete, and so there is a need for techniques that automatically assign GO functional categories to genes based on integration of available data. The present work describes one such technique, that uses a combination of sequence similarity and a similarity measure based on mutual information applied to cross-experiment microarray gene expression analysis. First of all, in order to test the relevance of sequence similarity for gene function inference, similarity searches of genes belonging to the same GO (from here on we will use “GO” as a shorthand for “GO category”, as well as for the “Gene Ontology” as a whole) were done across the human genome. A BLAST attachment value (BAV) for each GO was defined as the sum of the e-value exponents found between pairs of genes in the GO, divided by the sum of all e-value exponents found between genes in the GO and genes outside the GO. Next, to assess the “expression based similarity” of human genes, we used a dataset (GDS181) from GEO, a gene expression and molecular data repository maintained by the NCBI, providing gene expression profiles from 85 different tissues, organs, and cell lines in the normal physiological state. The dataset contains 12,625 probes, and we used 9,725 of them associated to genes with identifiable GO relationships. For each gene in the dataset, we calculated the Mutual Information (MI) between its expression values measured across all tissues and the corresponding values for the other genes. In order to calculate MI, the gene expression values were discretized, meaning that each one was replaced by one of K symbols. The symbol replacing an expression value was calculated by first normalizing the values into [0,1], and then partitioning this interval into K equally sized subintervals. The normalization was done on a per-gene basis. After experimenting with several different values of K, a value of K=3 was chosen for all further experiments. Using a similar procedure to the one used for calculating the BAV, a MI attachment value (MiAV) was obtained. For each GO, the MiAV was defined as the sum of the MI expression values found for all pairs of genes in the GO, divided by the sum of all MI values between genes in the GO and genes outside the GO. Then, our gene function inference (GFI) process proceeds as follows. Given a gene for which one wants to know the function, one begins by comparing it with all other genes, using both BLAST and expression data. Then, given a GO, one may calculate the values Bs and Ms, representing the maximum similarity found between the query gene and any gene inside the GO, using BLAST or MI, respectively. Those values plus attachments are used in the following equation for estimating the pertinence of a gene to a GO:

We posit that, given the current state of development of cognitive science, the greatest synergies between this field and artificial intelligence arise when one adopts a high level of abstraction. On the one hand, we suggest, cognitive science embodies some interesting, potentially general principles regarding cognition under limited resources, and AI systems that violate these principles should be treated with skepticism. But on the other hand, attempts to precisely emulate human cognition in silicon are hampered by both their ineffectiveness at exploiting the power of digital computers, and the current paucity of algorithm-level knowledge as to how human cognition takes place. We advocate a focus on artificial general intelligence design. This

Welcome to the HLT-NAACL'06 BioNLP Workshop, Linking Natural Language Processing and Biology: Towards Deeper Biological Literature Analysis. The late 1990s saw the beginning of a trend towards significant growth in the area of biomedical language processing, and in particular in the use of natural language processing techniques in the molecular biology and related computational bioscience domains. The figure below gives an indication of the amount of recent activity in this area: it shows the cumulative number of documents returned by searching PubMed, the premiere repository of biomedical scientific literature, with the query ((natural language processing) OR (text mining)) AND (gene OR protein), limiting the search by year for every year from 1999 through 2005: the three papers in 1999 had grown to 227 by the end of 2005. Significant challenges to biological literature exploitation remain, in particular for such biological problem areas as automated function prediction and pathway reconstruction and for linguistic applications such as relation extraction and abstractive summarization. In light of the nature of these remaining challenges, the focus of this workshop was intended to be applications that move towards deeper semantic analysis. We particularly solicited work that addresses relatively under-explored areas such as summarization and question-answering from biological information. Papers describing applications of semantic processing technologies to the biology domain were especially invited. That is, the primary topics of interest were applications which require deeper linguistic analysis of the biological literature. We also solicited papers exploring issues in porting NLP systems originally constructed for other domains to the biology domain. What makes the biology domain special? What hurdles must be overcome in performing linguistic analysis of biological text? Are any special linguistic or knowledge resources required, beyond a domain-specific lexicon? What relations in biological text are most interesting to biologists, and hence should be the focus of our future efforts? The workshop received 31 submissions: 29 full-paper submissions, and two poster submissions. A strong program committee, representing BioNLP researchers in North America, Europe, and Asia, provided thorough reviews, resulting in the acceptance of eleven full papers and nineteen posters, for an acceptance rate for full papers of 38% (11/29), which we believe made this one of the most competitive BioNLP workshop or conference sessions to date. A notable trend in the accepted papers is that only one of them was on the topic of entity identification. The subject areas of the papers presented at BioNLP'06 included an exceptionally wide range of topics: question-answering, computational lexical semantics, information extraction, entity normalization, semantic role labelling, image classification, and syntactic aspects of the sublanguage of molecular biology

The intent of this workshop was to bring researchers in text processing in the bioinformatics and biomedical domains together to discuss how techniques from natural language processing and information retrieval can be exploited to address biological information needs. Credit for its successes in reaching that goal is due entirely to the authors of the papers and posters presented in this volume and to the exceptional program committee.

We describe an extension of the Bayesian Optimization Algorithm (BOA), a probabilistic model building genetic algorithm, to the domain of program tree evolution. The new system, BOA programming (BOAP), improves significantly on previous probabilistic model building genetic programming (PMBGP) systems in terms of the articulacy and open-ended flexibility of the models learned, and hence control over the distribution of instances generated. Innovations include a novel tree representation and a generalized program evaluation scheme.

The Novamente AI Engine is briefly reviewed. The overall architecture is unique, drawing on system-theoretic ideas regarding complex mental dynamics and associated emergent patterns. We describe how these are facilitated by a novel knowledge representation which allows diverse cognitive processes to interact effectively. We then elaborate the two primary cognitive algorithms used to construct these processes: probabilistic term logic (PTL), and the Bayesian Optimization Algorithm (BOA). PTL is a highly flexible inference framework, applicable to domains involving uncertain, dynamic data, and autonomous agents in complex environments. BOA is a population-based optimization algorithm which can incorporate prior knowledge. While originally designed to operate on bit strings, our extended version also learns programs and predicates with variable length and tree-like structure, used to represent actions, perceptions, and internal state. We detail some of the specific dynamics and structures we expect to emerge through the interaction of the cognitive processes, outline our approach to training the system through experiential interactive learning, and conclude with a description of some recent results obtained with our partial implementation, including practical work in bioinformatics, natural language processing, and knowledge discovery.

This volume presents a collection of essays that all share a common concern with time, process and consciousness. The chapters represent a variety of different perspectives and the authors span the disciplines of psychology, mathematics, physics and psychiatry.

A novel, modular connectionist architecture for grammar induction is proposed, the "Markovian Language Network" (MLN). Based on Zellig Harris's (1989) theory of syntax, MLN would seem to be a plausible biological and psychological model. Modules for extraction of Markov information and for categorization are interlinked and are used for multiple purposes. MLN has not yet been fully implemented, but simple experiments have been conducted with a hybrid system. MLN is seen as a first step toward a complete connectionist implementation of Harrisian linguistics.

The IMGA (iterated mean genetic algorithm), a real nonlinear iteration which approximates the GA, is introduced. The IMGA approximates the GA arbitrarily closely as population size approaches infinity. The eigenvalues of the Jacobian of the IMGA are evaluated for the case of a globally optimal population. This computation shows that, in the case of no mutation, and a fitness function with a unique global maximum, a globally optimal population is an attractor for the IMGA. A bound on the asymptotic rate of convergence of the IMGA is given.

In the Webmind network-based self-organizing knowledge management system, symbols are grounded by the genetic-algorithm evolution of automata mapping nonlinguistic data to patterns of linguistic reference associated with symbols. This process is the core of the Webmind system's creation of meaning from its own experience, leading to the philosophical conclusion that, in Webmind at any rate, semiotics supports autonomy via the social creation of self, whereas autonomy supports semiotics by supplying a coherent body of information to serve as raw material for the evolution of symbol groundings.

第１章 少年時代（１９０１‐１９１３） 第２章 高校落第生（１９１３‐１９２２） 第３章 結晶学に魅せられて（１９２２‐１９２５） 第４章 化学結合の本質（１９２６‐１９３５） 第５章 タンパク質、政治そしてパスポート（１９３５‐１９５４） 第６章 死の灰（１９５４‐１９６０） 第７章 平和運動のチャンピオン（１９６０‐１９６６） 第８章 万能薬ビタミンＣ（１９６６‐１９９０） 第９章 ひとり海辺で（１９９１‐１９９４）

The psynet model, a complexity theory of mind, is summarized and discussed. The mind is viewed as an autopoletic system of pattern/processes, which gives rise to a hierarchy of attractor structures, culminating in a “dual network” attractor consisting of superposed hierarchical and associative networks. Learning is accomplished by genetic programming on process trees; persistent beliefs and symbolic information are achieved by stable systems of intercreating processes.

The psynet model, a complexity theory of mind, is summarized and discussed. The mind is viewed as an autopoletic system of pattern/processes, which gives rise to a hierarchy of attractor structures, culminating in a “dual network” attractor consisting of superposed hierarchical and associative networks. Learning is accomplished by genetic programming on process trees; persistent beliefs and symbolic information are achieved by stable systems of intercreating processes.

Journal Article Force of Nature: The Life of Linus Pauling. By Thomas Hager. (New York: Simon & Schuster, 1995 and Linus Pauling: A Life in Science and Politics. By Ted Goertzel and Ben Goertzel. New York: Basic Books, 1995 Get access Daniel J. Kevies Daniel J. Kevies California Institute of Technology, Pasadena, California Search for other works by this author on: Oxford Academic Google Scholar Journal of American History, Volume 83, Issue 3, December 1996, Pages 983–985, https://doi.org/10.2307/2945656 Published: 01 December 1996

Abstract Images of mind is an exciting book, well-written and wellorganized, but many of the connections the authors draw between PET scan results and more general psychological issues are somewhat strained.

Abstract In their natural condition, hydrocollapsible soils exhibit considerable strength. As they become wet, they exhibit considerable collapse, even in the absence of an applied load. The severity of this collapse depends on a variety of factors. One would like to be able to predict the severity of collapse from easily measurable properties of the soil - if not with absolute precision, at least in an approximate, probabilistic way. Using a hydrocollapse measurements database compiled from geotechnical reports on various clays, silts and sands found in the Las Vegas area, we apply linear discriminant analysis to determine the dependence of collapse percentage on four properties: moisture content, load, dry density, and depth. Using this analysis, we give a formula which estimates into which of three predetermined collapse ranges a given sample is most likely to fall, either based on all four properties or based on moisture content and dry density alone. We then derive a novel probabilistic classification rule, which estimates the probability that a soil sample falls into a given collapse range. Although quite simple, this rule uses much more of the information provided by discriminant analysis than does simple deterministic classification.

The chaotic dynamics of the map ϕ ( x ) = ( β x + α )(mod1) are studied using Parry′s β ‐expansion. It is shown that for 1 &lt; β &lt; 2, a ≥ 0, the number of periodic points of period n is 0( β n ).

A function f : {0,1,2, L , a } n → R is said to be uncorrelated if Prob[f(x) ≤ u ] = G ( u ). This paper studies the effectiveness of simulated annealing as a strategy for optimizing uncorrelated functions. A recurrence relation expressing the effectiveness of the algorithm in terms of the function G is derived. Surprising numerical results are obtained, to the effect that for certain parametrized families of functions { G c , c ∈ R }, where c represents the steepness of the curve G ′ ( u ), the effectiveness of simulated annealing increases steadily with c These results suggest that on the average annealing is effective whenever most points have very small objective function values, but a few points have very large objective function values.

(1991). The Rock Guitar Solo: From expression to simulation. Popular Music and Society: Vol. 15, No. 1, pp. 91-101.

The concept of pattern is introduced, formally defined, and used to analyze various measures of the complexity of finite binary sequences and other objects. The standard Kolmogoroff‐Chaitin‐Solomonoff complexity measure is considered, along with Bennett′s logical depth, Koppel′s sophistication′, and Chaitin′s analysis of the complexity of geometric objects. The pattern‐theoretic point of view illuminates the shortcomings of these measures and leads to specific improvements, it gives rise to two novel mathematical concepts‐‐orders of complexity and levels of pattern, and it yields a new measure of complexity, the structural complexity, which measures the total amount of structure an entity possesses.