The problem of expensive chunks and its solution by restricting expressiveness

Soar is an architecture for a system that is intended to be capable of general intelligence. Chunking, a simple experience-based learning mechanism, is Soar's only learning mechanism. Chunking creates new items of information, called chunks, based on the results of problem-solving and stores them in the knowledge base. These chunks are accessed and used in appropriate later situations to avoid the problem-solving required to determine them. It is already well-established that chunking improves performance in Soar when viewed in terms of the subproblems required and the number of steps within a subproblem. However, despite the reduction in number of steps, sometimes there may be a severe degradation in the total run time. This problem arises due toexpensive chunks, i.e., chunks that require a large amount of effort in accessing them from the knowledge base. They pose a major problem for Soar, since in their presence, no guarantees can be given about Soar's performance.In this article, we establish that expensive chunks exist and analyze their causes. We use this analysis to propose a solution for expensive chunks. The solution is based on the notion of restricting the expressiveness of the representational language to guarantee that the chunks formed will require only a limited amount of accessing effort. We analyze the tradeoffs involved in restricting expressiveness and present some empirical evidence to support our analysis.

[1]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[2]  M. Garey Johnson: computers and intractability: a guide to the theory of np- completeness (freeman , 1979 .

[3]  Allen and Rosenbloom Paul S. Newell,et al.  Mechanisms of Skill Acquisition and the Law of Practice , 1993 .

[4]  J. D. Uiiman Principles of database systems , 1982 .

[5]  Charles L. Forgy,et al.  The OPS83 report , 1984 .

[6]  Allen Newell,et al.  R1-Soar: An Experiment in Knowledge-Intensive Programming in a Problem-Solving Architecture , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Steven Minton,et al.  Selectively Generalizing Plans for Problem-Solving , 1985, IJCAI.

[8]  David E. Smith,et al.  Ordering Conjunctive Queries , 1985, Artif. Intell..

[9]  H. Levesque,et al.  Readings in Knowledge Representation , 1985 .

[10]  Elaine Kant,et al.  Programming expert systems in OPS5 , 1985 .

[11]  Nancy Martin,et al.  Programming Expert Systems in OPS5 - An Introduction to Rule-Based Programming(1) , 1985, Int. CMG Conference.

[12]  Daniel J. Scales Efficient matching algorithms for the Soar/OPS5 production system , 1986 .

[13]  John E. Laird,et al.  Mapping Explanation-Based Generalization onto Soar , 1986, AAAI.

[14]  Daniel P. Miranker TREAT: a better match algorithm for AI production systems , 1987, AAAI 1987.

[15]  Kemal Oflazer,et al.  Partitioning in parallel processing of production systems , 1987 .

[16]  David M. Steier CYPRESS-Soar: A Case Study in Search and Learning in Algorithm Design , 1987, IJCAI.

[17]  G. Weiderhold File organization for database design , 1987 .

[18]  Gerald DeJong,et al.  An Explanation-based Approach to Generalizing Number , 1987, IJCAI.

[19]  Anoop Gupta Parallelism in production systems , 1987 .

[20]  Allen Newell,et al.  SOAR: An Architecture for General Intelligence , 1987, Artif. Intell..

[21]  Jack Mostow,et al.  PROLEARN: Towards a Prolog Interpreter that Learns , 1987, AAAI.

[22]  Allen Newell,et al.  Some Chunks Are Expensive , 1988, ML.

[23]  William W. Cohen Generalizing Number and Learning from Multiple Examples in Explanation Based Learning , 1988, ML.

[24]  Richard M. Keller,et al.  Defining Operationality for Explanation-Based Learning , 1987, Artificial Intelligence.

[25]  Anoop Gupta,et al.  Suitability of Message Passing Computers for Implementing Production Systems , 1988, AAAI.

[26]  Jaime G. Carbonell,et al.  Learning effective search control knowledge: an explanation-based approach , 1988 .

[27]  Anoop Gupta,et al.  Comparison of the RETE and TREAT production matchers for soar (A summary) , 1988, AAAI 1988.

[28]  Toru Ishida,et al.  Optimizing Rules in Production System Programs , 1988, AAAI.

[29]  Shaul Markovitch,et al.  The Role of Forgetting in Learning , 1988, ML.

[30]  Allen Newell,et al.  Soar/PSM-E: investigating match parallelism in a learning production sytsem , 1988, PPoPP 1988.

[31]  Raymond J. Mooney,et al.  The Effect of Rule Use on the Utility of Explanation-Based Learning , 1989, IJCAI.

[32]  Shaul Markovitch,et al.  Information Filters and Their Implementation in the SYLLOG System , 1989, ML.

[33]  Oren Etzioni,et al.  Explanation-Based Learning: A Problem Solving Perspective , 1989, Artif. Intell..

[34]  Paul P. Maglio,et al.  Approximating Learned Search Control Knowledge , 1989, ML.

[35]  John E. Laird,et al.  Symbolic architectures for cognition , 1989 .

[36]  Allen Newell,et al.  A Problem Space Approach to Expert System Specification , 1989, IJCAI.

[37]  Jaime G. Carbonell,et al.  Towards a General Framework for Composing Disjunctive and Iterative Macro-operators , 1989, IJCAI.

[38]  M. Posner Foundations of cognitive science , 1989 .

[39]  Shaul Markovitch,et al.  Utilization Filtering: A Method for Reducing the Inherent Harmfulness of Deductively Learned Knowledge , 1989, IJCAI.

[40]  Jude W. Shavlik,et al.  Acquiring Recursive Concepts with Explanation-Based Learning , 1989, IJCAI.

[41]  Milind Tambe,et al.  Eliminating Expensive Chunks by Restricting Expressiveness , 1989, IJCAI.

[42]  Russell Greiner,et al.  Incorporating Redundant Learned Rules: A Preliminary Formal Analysis of EBL , 1989, IJCAI.

[43]  Milind Tambe,et al.  A Frameworkfor Investigating Production System Formulations with Polynomially Bounded Match , 1990, AAAI.

[44]  Krzysztof R. Apt,et al.  Logic Programming , 1990, Handbook of Theoretical Computer Science, Volume B: Formal Models and Sematics.

[45]  Steven Minton,et al.  Quantitative Results Concerning the Utility of Explanation-based Learning , 1988, Artif. Intell..

[46]  Richard Reviewer-Granger Unified Theories of Cognition , 1991, Journal of Cognitive Neuroscience.

[47]  Charles L. Forgy,et al.  Rete: a fast algorithm for the many pattern/many object pattern match problem , 1991 .

[48]  Allen Newell,et al.  Modeling human syllogistic reasoning in Soar , 1993 .

[49]  Allen Newell,et al.  The chunking of goal hierarchies: a generalized model of practice , 1993 .

[50]  Allen Newell,et al.  Varieties of learning in Soar: 1987 , 1993 .

[51]  Paul S. Rosenbloom,et al.  Applying problem solving and learning to diagnosis , 1993 .

[52]  Tom M. Mitchell,et al.  Explanation-Based Generalization: A Unifying View , 1986, Machine Learning.

[53]  Glenn A. Iba,et al.  A Heuristic Approach to the Discovery of Macro-Operators , 1989, Machine Learning.

[54]  Allen Newell,et al.  Chunking in Soar: The anatomy of a general learning mechanism , 1985, Machine Learning.

[55]  Gerald DeJong,et al.  Explanation-Based Learning: An Alternative View , 2005, Machine Learning.