Structural representation and reasoning in a hybrid cognitive architecture

Psychologically and neurobiologically plausible models of knowledge often must make a difficult choice between distributed and localist representation. Distributed representation can be flexible and hold up well to noisy data, but localist models allow for structured knowledge to be represented unambiguously and reasoned over in rigorous, transparent fashion. We present a way of representing knowledge within the hybrid cognitive architecture CLARION. Our system allows both structured knowledge and distributed knowledge to synergistically coexist while remaining within the limits defined by CLARION'S dual-process framework. After showing how our system can allow more complex knowledge structures to arise, we describe algorithms that use such structures to model many types of reasoning, including: analogical reasoning, deductive reasoning, moral reasoning, and more. We place the structural knowledge afforded CLARION within a formal hierarchy of expressivity for such knowledge, and discuss implications of this work.

[1]  J. Hummel,et al.  Exploring the Role of Analogico-Deductive Reasoning in the Balance-Beam Task , 2012 .

[2]  Selmer Bringsjord,et al.  The Cambridge Handbook of Computational Psychology: Declarative/Logic-Based Cognitive Modeling , 2008 .

[3]  Ron Sun,et al.  Duality of the mind - a bottom-up approach toward cognition , 2001 .

[4]  Naveen Sundar Govindarajulu,et al.  How Models of Creativity and Analogy Need to Answer the Tailorability Concern , 2015 .

[5]  Ron Sun,et al.  Moral Judgment, Human Motivation, and Neural Networks , 2013, Cognitive Computation.

[6]  Markus Werning,et al.  Compositionality and Biologically Plausible Models , 2009 .

[7]  John E. Hummel,et al.  Relational Reasoning in a Neurally Plausible Cognitive Architecture , 2005 .

[8]  Xi Zhang,et al.  Accounting for a variety of reasoning data within a cognitive architecture , 2006, J. Exp. Theor. Artif. Intell..

[9]  Selmer Bringsjord,et al.  Analogico-Deductive Generation of Gödel's First Incompleteness Theorem from the Liar Paradox , 2013, IJCAI.

[10]  Dedre Gentner,et al.  Structure-Mapping: A Theoretical Framework for Analogy , 1983, Cogn. Sci..

[11]  R. Sun.  Connectionist Models of Rule-Based ReasoningRon , 1991 .

[12]  A. Tversky.  Features of Similarity , 1977 .

[13]  Ron Sun,et al.  Robust Reasoning: Integrating Rule-Based and Similarity-Based Reasoning , 1995, Artif. Intell..

[14]  John E. Hummel,et al.  Higher-Level Cognitive Functions and Connectionist Modeling. Relational Reasoning in a Neurally-Plausible Cognitive Architecture: An Overview of the LISA Project. , 2003 .

[15]  R. Sun.  Accounting for Similarity-Based Reasoning within a Cognitive Architecture , 2022 .

[16]  Selmer Bringsjord,et al.  Is the connectionist-logicist clash one of AI's wonderful red herrings? , 1991, J. Exp. Theor. Artif. Intell..

[17]  Marco Dorigo,et al.  Ant colony optimization theory: A survey , 2005, Theor. Comput. Sci..

[18]  L. Shastri,et al.  From Simple Associations to Systemic Reasoning: A Connectionist Representation of Rules, Variables and Dynamic Bindings , 1990 .

[19]  Jonathan Evans.  Dual-processing accounts of reasoning, judgment, and social cognition. , 2008, Annual review of psychology.

[20]  Christine Solnon,et al.  Ant Algorithm for the Graph Matching Problem , 2005, EvoCOP.

[21]  Geoffrey E. Hinton,et al.  Distributed Representations , 1986, The Philosophy of Artificial Intelligence.

[22]  Ron Sun,et al.  A Microfeature Based Approach Towards Metaphor Interpretation , 1995, IJCAI.

[23]  Artur S. d'Avila Garcez,et al.  A Connectionist Computational Model for Epistemic and Temporal Reasoning , 2006, Neural Computation.

[24]  R. Sun.  Desiderata for cognitive architectures , 2004 .