AGI

1. NEW: Propositional Probabilistic Graphical Models Δ (TeXmacs source Δ)

1. Adaptive Logics for Reasoning Systems

1. Frequency and/or Uncertainty Logics: Non Axiomatic Logic, Probabilistic Logic Networks (to come)

1. Probabilistic Modeling and Probabilistic Logics
   1. OLD: Markov Logic Networks Δ
   2. NEW: Propositional Probabilistic Graphical Models Δ (TeXmacs source Δ)
   3. to come: Relational Probabilistic Models and Logics

• A Quick and Thorough Introduction to ACT-R 6 Δ (odp source Δ)

1. Frequency and/or Uncertainty Logics: Non Axiomatic Logic, Markov Logic Networks Δ, Probabilistic Logic Networks (to come)

1. Inductive (Logic) Programming

• A Quick and Thorough Introduction to ACT-R 6 Δ (odp source Δ), NEW

• A Quick and Thorough Introduction to ACT-R 6? (odp source?), NEW

1. General Game Playing Δ (TeXmacs source Δ)

• Automated Language Acquisition Δ

test

1. RL_Ch6_Evolutionary_modular.pdf Δ

1. RL_Ch7_Hierarchical_RL.pdf Δ

1. General Game Playing Δ (TeXmacs source Δ) (NEW: a couple of pages about transfer learning)

• OpenCog Prime Δ

1. Adaptive and Probabilistic Logics for Reasoning Systems
   1. Adaptive (or Defeasible) Logics and OSCAR Δ (TODO: complete the notes about OSCAR)
   2. Frequency and/or Uncertainty Logics: Non Axiomatic Logic, Markov Logic Networks, Probabilistic Logic Networks Δ (to come)

• OpenCog Prime Δ (work in progress)

Notes:

• A glimpse on Warren Smith’s idea of IQ test Δ (source Δ)

1. Estimation of Distribution Algorithms and Genetic Programming Δ (TeXmacs source Δ)

1. Cognitive Architectures
   • Agents, Cognitive Architectures and Artificial General Intelligence Projects Δ (OpenOffice presentation source Δ)
   • OpenCog Prime? (work in progress)

• We build semantics in both representational and logic aspects based on the notion of agent specific rewards/motivations.

• Ontological Semantics Δ

• Multi-layered Semantic Networks Δ (replaces a subchapter above)
• Ontological Semantics Δ (work in progress)

1. Adaptive and Probabilistic Logics for Reasoning Systems Δ
   • some work in progress (PLN)

1. The Representation and Acquisition of Concepts Δ (to come later)

1. Cognitive Architectures Δ (to come soon)

1. Spreading Activation: memory retrieval, distributed reasoning, action selection, probabilities Δ (work in progress) approaches based on spreading activation mechanism or strong biological inspirations

• Multi-layered Semantic Networks Δ (subchapter to be included above)

1. Adaptive and Probabilistic Logics for Reasoning Systems Δ (progress on PLN will be resumed when some OpenCog documentation and code is released and the PLN book gets published)

1. Spreading Activation: memory retrieval, distributed reasoning, action selection, probabilities? (work in progress) approaches based on spreading activation mechanism or strong biological inspirations

Considered:

1. Values and Others: Grounding Agents in Game Semantics Δ (to come)
   • We build semantics in both representational and logic aspects based on the notion of agent specific rewards/motivations, interaction protocols etc.

• SOAR
• ACT-R
• OSCAR

• Multi-layered Semantic Networks Δ

1. Multi-layered Semantic Networks Δ

1. Knowledge Representation and Language Δ (TeXmacs source Δ)

• This chapter builds semantics in both representational and logic aspects based on the notion of agent specific rewards/motivations, interaction protocols etc.

• This chapter builds semantics in both representational and logic aspects based on the notion of agent/actor specific rewards, interaction protocols etc.

1. Knowledge Representation and Language Δ (in progress)

1. Values and Others: Grounding Agents in Game Semantics Δ (to come)
   • This chapter builds semantics in both representational and logic aspects based on the notion of agent/actor specific rewards

• based mostly on The Representation and Acquisition of Concepts. Psychology 747, Section 4097 - Fall 2003, but also “Conceptual Spaces. Geometry of thought” and more
• perhaps will be interrelated with “Automatic Programming” and some data mining ideas

I’ve moved KR before logics to introduce representation means of representation-specific reasoning systems (NARS and PLN) there.

• I’ve presented OSCAR and comments on John Pollock’s theory supported by his slides and article figures, I also introduced “propositional” NARS

1. Knowledge Representation (and Language) Δ (in progress)

1. Adaptive and Probabilistic Logics for Reasoning Systems Δ (progress on chapters beyond the finished “adaptive logics” will be resumed somewhere in March)
   • I’ve presented OSCAR and comments on John Pollock’s theory from his slides and article figures, I also introduced “propositional” NARS

• Perhaps I’ll merge it with the previous KR subject (that was my initial plan, but the concepts subject is vast enough…) Perhaps I’ll merge some and leave some here.
• based mostly on The Representation and Acquisition of Concepts. Psychology 747, Section 4097 - Fall 2003, but also “Conceptual Spaces. Geometry of thought”

1. Knowledge Representation (and Language) Δ (to come)
2. The Representation and Acquisition of Concepts Δ (to come)
   1. Perhaps I’ll merge it with the previous KR subject (that was my initial plan, but the concepts subject is vast enough…) Perhaps I’ll merge some and leave some here.
   2. based mostly on The Representation and Acquisition of Concepts. Psychology 747, Section 4097 - Fall 2003, but also “Conceptual Spaces. Geometry of thought”

1. Adaptive and Probabilistic Logics for Reasoning Systems Δ (in progress)

1. Adaptive Logics and Reasoning Systems Δ (in progress)
2. Knowledge Representation Δ (to come)

1. Knowledge Representation Δ (in progress)

• Proceedings of the AGIRI Workshop 2006 (Editors: Ben Goertzel, Pei Wang), IOS Press
• AGI-08. Proceedings of the First Conference on Artificial General Intelligence, (Editors: Pei Wang, Ben Goertzel, Stan Franklin), IOS Press

1. Cognitive Architectures Δ (to come)

• Overview of the Texai project (Steve Reed)

Main stream notes:

Auxiliary stream notes:

1. A glimpse on Warren Smith’s idea of IQ test Δ (source Δ)
2. Agents, Cognitive Architectures and Artificial General Intelligence Projects Δ (OpenOffice presentation source Δ)
3. Learning the structure and parameters of graphical models (Bayesian and neural networks) Δ (to come)
4. Estimation of Distribution Algorithms and Genetic Programming Δ (TeXmacs source Δ)
5. Automatic Deductive and Inductive Programming Δ (to come)

Overall course focus: (hmm…)

1. Techniques of Reinforcement Learning Δ (TeXmacs source Δ)
2. General Game Playing Δ (TeXmacs source Δ)

1. Adaptive Logics and Reasoning Systems Δ (to come)

1. (to come)

• “The Cambridge Handbook of Thinking and Reasoning”, Keith Holyoak, Robert Morrison (editors), Cambridge University Press, 2005

Some video lectures (currently not well selected):

• Reinforcement Learning and Artificial Intelligence “RLAI research is research directed toward the long-standing goals of AI (understanding the mind, reproducing human abilities) and is based on reinforcement learning ideas (learning from and while interacting with the world).”

Video lectures:

• IBM Research’s Almaden Institute Conference on Cognitive Computing
• Model-based Bayesian RL and related lectures from ICML-07 Bayesian RL Tutorial (site)

• SHRUTI “From Simple Associations to Systematic Reasoning”

• OpenCog

Video lectures:

• IBM Research’s Almaden Institute Conference on Cognitive Computing
• Model-based Bayesian RL and related lectures from ICML-07 Bayesian RL Tutorial (site)

1. “Neuro-Dynamic Programming: An Overview,” A slide presentation by Dimitri P. Bertsekas

1. General Game Playing Δ (to come)

• “Knowledge Representation and the Semantics of Natural Language”, Hermann Helbig, 2006, Cognitive Technologies series at Springer

1. Techniques of Reinforcement Learning Δ (TeXmacs source Δ)

1. Introduction to Reinforcement Learning Δ (to come)
2. Learning the structure and parameters of graphical models (Bayesian and neural networks) Δ (to come)
3. Estimation of Distribution Algorithms and Genetic Programming Δ (TeXmacs source Δ)
4. Automatic Deductive and Inductive Programming Δ (to come)
5. Adaptive Logics and Reasoning Systems Δ (to come)
6. Knowledge Representation Δ (to come)

1. Universal Artificial Intelligence: universal induction, exhaustive program search and reinforcement learning algorithms Δ (TeXmacs source Δ)

1. Universal Artificial Intelligence: universal induction, exhaustive program search and reinforcement learning algorithms Δ (TeXmacs source Δ)

• Mathematical definition of “intelligence” (and consequences), Warren D. Smith, June 2006

Places:

• Artificial General Intelligence Research Institute
• Association for Uncertainty in Artificial Intelligence
• Adaptive Logics Home Page

Attic

Outdated plan:

1. Information, distributions, programs, intelligence.
   1. Shannon information and Kolmogorov information, measures of complexity.
   2. Decision and control theory topics. Markov decision processes, reinforcement learning (Q-learning, SARSA etc.).
      1. SAIL and Dav: robots that learn “from scratch”.
   3. “General algorithmic intelligence” AIXI.
   4. Self improving programs: “Goedel Machine”. “Verificationist” program synthesis.
   5. Graphical probability models.
      1. Bayesian networks.
      2. Hierarchical Temporal Memory from Numenta.
      3. Introduction to “estimation of distribution” algorithms.
   6. “Optimizationist” competent program synthesis: algorithm MOSES.
   7. Representing and learning concepts. PAC-learnability.
      1. Learning grammars.
      2. Higher order and recursive structure representation induction.
2. Logic in a dynamic world.
   1. Adaptive logics overview (circumscription, defeasible argumentation, belief revision, etc.) Intensional and term logics.
      1. From semantic networks to logic: system SNePS.
   2. Reasoning about probability and uncertainty.
      1. Game semantics for logics.
      2. Recursive probability models.
      3. Probabilistic term logic. “Two-dimensional” truth values: system NARS.
      4. “Probabilistic Logic Networks” in Novamente.
   3. Inductive probabilistic logic programming vel probabilistic logic learning.
3. Cognitive loop (in search for the “main()” of the artificial mind).
   1. Inference system as an agent: goals and activations. (SNePS, NARS)
   2. Cognitive loop in LIDA.
   3. Concept formation and modeling of self. (Novamente)
4. Theory of mental development.
   1. Piagetan psychology.
   2. Mental development of an AGI.

Slides:

1. A glimpse on Warren Smith’s idea of IQ test Δ (source Δ)
2. Universal Artificial Intelligence: universal induction, exhaustive program search and reinforcement learning algorithms Δ (to come)

Attic

More online reading:

• A Working Hypothesis for General Intelligence, Eric B. Baum, October 2006
• Mathematical definition of “intelligence” (and consequences), Warren D. Smith, May 2006

• slides: Probabilistic Logic Learning Taneli Mielikinen 2005

Major offline reading (available to me):

• “Artificial General Intelligence”, Ben Goertzel, Cassion Pennachin (editors), 2007, Cognitive Technologies series at Springer
• “Rigid Flexibility. The Logic of Intelligence”, Pei Wang, 2006, Applied Logic series at Springer
• “Universal Artificial Intelligence. Sequential Decisions based on Algorithmic Probability”, Marcus Hutter, 2005, Texts in Theoretical Computer Science series at Springer

Major online reading:

• Thinking about Acting by John Pollock

• slides: Probabilistic Logic Learning Taneli Mielikäinen 2005

• NIPS 2005 workshop. Towards human-level AI? (with slides)

• NIPS 2005 workshop. Towards human-level AI?

• slides: Probabilistic Logic Learning Taneli Mielikäinen 2005

• Proceedings of the AGIRI Workshop 2006

• slides: Probabilistic Logic Learning Taneli Mielikäinen 2005

1. Adaptive logics overview (circumscription, defeasible argumentation, belief revision, etc.) Intensional and term logics.

1. “Optimizationist” competent program synthesis: algorithm MOSES.

1. Higher order and recursive structure representation induction.

1. Graphical probability models.
   1. Bayesian networks.

1. Introduction to “estimation of distribution” algorithms.

1. Hierarchical Temporal Memory from Numenta.
2. “Optimizationist” competent program synthesis: algorithm MOSES.

1. Learning grammars.

1. Adaptive logics overview (circumscription, belief revision, etc.) Intensional and term logics.

1. Probabilistic term logic. “Two-dimensional” truth values: system NARS.

1. Inference system as an agent: goals and activations. (SNePS, NARS)

1. SAIL and Dav: robots that learn “from scratch”.

• Machine Learning Lecture Notes, a MIT course

• Novamente

• NARS

• LEVELS OF ORGANIZATION IN GENERAL INTELLIGENCE Eliezer S. Yudkowsky

1. Adaptive and term logics overview (intensionality, circumscription, belief revision, etc.)

• Adaptive Logics Home Page

Obligatory reading:

• Reinforcement Learning: An Introduction Richard S. Sutton and Andrew G. Barto

Reviews: Artificial General Intelligence: A Gentle Introduction by Pei Wang

Places:

• Artificial General Intelligence Research Institute
• Association for Uncertainty in Artificial Intelligence

• slides: Probabilistic Logic Learning Taneli Mielikäinen 2005

1. Decision and control theory topics. Markov decision processes, reinforcement learning (Q-learning, SARSA etc.).

1. Inference system as an agent: goals and activations. (NARS, Novamente)

1. From semantic networks to logic: system SNePS.

1. Concept formation and modeling of self. (Novamente)

1. Concept formation and modeling of self.

1. Inference system as an agent: goals and activations.

1. From inference to synthesis and back: system Novamente.

Overall course focus:

1. Reinforcement learning.
2. Concept formation and program synthesis.
3. Adaptive and probabilistic logics.
4. Mental development theory.

• Reinforcement Learning: An Introduction Richard S. Sutton and Andrew G. Barto

Tentative plan:

1. Information, distributions, programs, intelligence.
   1. Shannon information and Kolmogorov information, measures of complexity.
   2. Decision and control theory topics. Markov decision processes, reinforcement learning (Q- and TD- learning, SARSA etc.).
   3. “General algorithmic intelligence” AIXI.
   4. Self improving programs: “Goedel Machine”. “Verificationist” program synthesis.
   5. Bayesian networks. Introduction to “estimation of distribution” algorithms.
   6. “Optimizationist” competent program synthesis: algorithm MOSES.
   7. Representing and learning concepts. PAC-learnability.
   8. Learning grammars.
2. Logic in a dynamic world.
   1. Adaptive logics overview (intensional, term logic, circumscription, belief revision, etc.)
      1. From semantic networks to logic.
   2. Reasoning about probability and uncertainty.
      1. Game semantics for logics.
      2. Recursive probability models.
      3. Probabilistic term logic.
      4. “Two-dimensional” truth values: system NARS.
      5. “Probabilistic Logic Networks” in Novamente.
   3. Inductive probabilistic logic programming vel probabilistic logic learning.
   4. From inference to synthesis and back: system Novamente.
3. Cognitive loop (in search for the “main()” of the artificial mind).
   1. Cognitive loop in LIDA.
   2. Cognitive loop (in broad sense) in Novamente.
4. Theory of mental development.
   1. Piagetan psychology.
   2. Mental development of an AGI.

• slides: Probabilistic Logic Learning Taneli Mielikäinen 2005

Some useful theory and mechanisms:

• Probabilistic Reasoning for Complex Systems, A.J. Pfeffer. PhD Thesis, 2000

• Logical Markov Decision Programs (2003) Kristian Kersting, Luc De Raedt

• Probabilistic Inductive Logic Programming, Luc De Raedt and Kristian Kersting, 2004

Important theory and mechanisms:

• slides: Probabilistic Logic Learning Taneli Mielikäinen 2005

• Semantics and Inference for Recursive Probability Models Avi Pfeffer, Daphne Koller, 2000

More:

• Probabilistic Logic Learning (2004) Luc De Raedt, Kristian Kersting
• A Revisionist View of Blackboard Systems Norman Carver, 1997

Some random papers:

Some technical papers:

• Norman Carver. A Revisionist View of Blackboard Systems 1997
• Logical Markov Decision Programs (2003) Kristian Kersting, Luc De Raedt

Architectures / projects:

• Novamente (główny architekt: Ben Goertzel) patrz też Is AI Engineering the Shortest Path to a Positive Singularity?
• LIDA
• NARS (Pei Wang)
• PolyScheme
• SNePS
• LEVELS OF ORGANIZATION IN GENERAL INTELLIGENCE Eliezer S. Yudkowsky (SIAI) (zobacz też Artificial Intelligence as a Precise Art Δ)
• AdaptiveAI
• Project Joshua Blue (Bootstrapping semantics in an autonomic computing system)
• Ai
• SAIL (Self-organizing Autonomous Incremental Learner)

Other links:

• Essentials of General Intelligence: The direct path to AGI by Peter Voss
• AGIRI AGI Research Institute wiki
• Marcus Hutter / Towards a Universal Theory of Artificial Intelligence based on Algorithmic Probability and Sequential Decisions
• Juergen Schmidhuber / The New AI: General & Sound & Relevant for Physics
• Site by Włodzisław Duch
• Confabulation Theory
• Logica Universalis
• The Emotion Machine