Exploiting persistent mappings in cross-domain analogical learning of physical domains

Cross-domain analogies are a powerful method for learning new domains. This paper extends the Domain Transfer via Analogy (DTA) method with persistent mappings, correspondences between domains that are incrementally built up as a cognitive system gains experience with a new domain. DTA uses analogies between pairs of textbook example problems, or worked solutions, to create a domain mapping between a familiar and a new domain. This mapping enables the initialization of a new domain theory. Another analogy is then made between the domain theories themselves, providing additional conjectures about the new domain. After these conjectures are verified, the successful mappings are stored as persistent mappings to constrain future analogies between the domains. We show that DTA plus persistent mappings enables a Companion, the first structure mapping cognitive architecture, to learn the equation schemas and control knowledge necessary to solve problems in three domains (rotational mechanics, electricity, and heat) by analogy with linear mechanics. We provide a detailed analysis categorizing transfer failures. As with people, the most difficult step in cross-domain analogy is identifying an appropriate example. Once an analogous example has been found, DTA successfully transfers the domain knowledge necessary to solve the problem in the new domain 78% of the time. Furthermore, we illustrate how persistent mappings assist in retrieval of analogous examples and overcoming two types of mapping failures.

[1]  Brian Falkenhainer,et al.  Learning from physical analogies: A study in analogy and the explanation process , 1989 .

[2]  D. Gentner.  Structure‐Mapping: A Theoretical Framework for Analogy* , 1983 .

[3]  Kenneth D. Forbus,et al.  Companion Cognitive Systems: Design Goals and Lessons Learned So Far , 2009, IEEE Intelligent Systems.

[4]  Kenneth D. Forbus,et al.  Building Problem Solvers , 1993 .

[5]  John E. Laird,et al.  Human-Level AI's Killer Application: Interactive Computer Games , 2000, AI Mag..

[6]  John R. Anderson,et al.  MACHINE LEARNING An Artificial Intelligence Approach , 2009 .

[7]  Kenneth D. Forbus,et al.  Analogical Learning in a Turn-Based Strategy Game , 2007, IJCAI.

[8]  John E. Laird,et al.  Integrating Reinforcement Learning with Soar. , 2004 .

[9]  Moshe Burstein,et al.  Concept formatio n by incremental analogica l reasoning and debugging , 1983 .

[10]  Kenneth D. Forbus,et al.  MAC/FAC: A Model of Similarity-Based Retrieval , 1995, Cogn. Sci..

[11]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[12]  P ? ? ? ? ? ? ? % ? ? ? ? , 1991 .

[13]  Peter Stone,et al.  Value-Function-Based Transfer for Reinforcement Learning Using Structure Mapping , 2006, AAAI.

[14]  Richard S. Sutton,et al.  Introduction to Reinforcement Learning , 1998 .

[15]  Scott E Friedman.  Learning Causal Models via Progressive Alignment & Qualitative Modeling : A Simulation , 2008 .

[16]  K. Holyoak,et al.  Schema induction and analogical transfer , 1983, Cognitive Psychology.

[17]  Pat Langley,et al.  Structural Transfer of Cognitive Skills , 2007 .

[18]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[19]  Dedre Gentner,et al.  Why we’re so smart , 2003 .

[20]  John E. Laird,et al.  Soar-RL: integrating reinforcement learning with Soar , 2005, Cognitive Systems Research.

[21]  Kenneth D. Forbus,et al.  Making SME greedy and pragmatic , 2006 .

[22]  Kai-Uwe Kühnberger,et al.  Metaphors and heuristic-driven theory projection (HDTP) , 2006, Theor. Comput. Sci..

[23]  Angela Schwering,et al.  THE IMPACT OF GESTALT PRINCIPLES ON SOLVING GEOMETRIC PROPORTIONAL ANALOGIES , 2009 .

[24]  Douglas C. Giancoli.  Physics: Principles with Applications , 1999 .

[25]  Kenneth D. Forbus,et al.  The Roles of Similarity in Transfer: Separating Retrievability From Inferential Soundness , 1993, Cognitive Psychology.

[26]  P. Langley.  The cognitive systems paradigm , 2012 .

[27]  Yusuf Pisan,et al.  An integrated architecture for engineering problem-solving , 1998 .

[28]  Kenneth D. Forbus,et al.  Analogy and creativity in the works of Johannes Kepler , 1997 .

[29]  Johan de Kleer,et al.  Fundamentals of model-based diagnosis , 2003 .

[30]  Paul Thagard,et al.  A computational model of analogical problem solving , 1989 .

[31]  David W. Aha,et al.  Learning Continuous Action Models in a Real-Time Strategy Environment , 2008, FLAIRS.

[32]  Raymond J. Mooney,et al.  Mapping and Revising Markov Logic Networks for Transfer Learning , 2007, AAAI.

[33]  Kenneth D. Forbus,et al.  Learning Modeling Abstractions via Generalization , 2008 .

[34]  Brian Falkenhainer,et al.  The Structure-Mapping Engine: Algorithm and Examples , 1989, Artif. Intell..

[35]  Kenneth D. Forbus,et al.  Transfer Learning through Analogy in Games , 2011, AI Mag..

[36]  D. Gentner,et al.  Flowing waters or teeming crowds: Mental models of electricity , 1982 .

[37]  Angela Schwering,et al.  Analogy as Integrating Framework for Human-Level Reasoning , 2008, AGI.

[38]  Kenneth D. Forbus,et al.  Domain transfer via cross-domain analogy , 2009, Cognitive Systems Research.

[39]  Kenneth D. Forbus,et al.  Measuring the Level of Transfer Learning by an AP Physics Problem-Solver , 2007, AAAI.

[40]  Tolga Könik,et al.  Achieving Far Transfer in an Integrated Cognitive Architecture , 2008, AAAI.

[41]  Matthew E. Taylor.  Transfer in Reinforcement Learning Domains , 2009, Studies in Computational Intelligence.

[42]  H. H. Richardson,et al.  Introduction to system dynamics , 1967 .

[43]  Kenneth D. Forbus,et al.  Exploring analogy in the large , 2000 .

[44]  Michael J. Witbrock,et al.  An Introduction to the Syntax and Content of Cyc , 2006, AAAI Spring Symposium: Formalizing and Compiling Background Knowledge and Its Applications to Knowledge Representation and Question Answering.

[45]  Jim Davies,et al.  Visual Analogy in Problem Solving , 2001, IJCAI.

[46]  Kenneth D. Forbus,et al.  Using analogical model formulation with sketches to solve Bennett Mechanical Comprehension Test problems , 2011, J. Exp. Theor. Artif. Intell..