Learning at the knowledge level

When Newell introduced the concept of the knowledge level as a useful level of description for computer systems, he focused on the representation of knowledge. This paper applies the knowledge level notion to the problem of knowledge acquisition. Two interesting issues arise. First, some existing machine learning programs appear to be completely static when viewed at the knowledge level. These programs improve their performance without changing their ‘knowledge’. Second, the behavior of some other machine learning programs cannot be predicted or described at the knowledge level. These programs take unjustified inductive leaps. The first programs are called symbol level learning (SLL) programs; the second, nondeductive knowledge level learning (NKLL) programs. The paper analyzes both of these classes of learning programs and speculates on the possibility of developing coherent theories of each. A theory of symbol level learning is sketched, and some reasons are presented for believing that a theory of NKLL will be difficult to obtain.

[1]  J. Armstrong.  Knowledge and Belief , 1953 .

[2]  Arthur L. Samuel,et al.  Some Studies in Machine Learning Using the Game of Checkers , 1967, IBM J. Res. Dev..

[3]  John McCarthy,et al.  Programs with common sense , 1960 .

[4]  E. Mark Gold,et al.  Language Identification in the Limit , 1967, Inf. Control..

[5]  Ryszard S. Michalski,et al.  On the Quasi-Minimal Solution of the General Covering Problem , 1969 .

[6]  Richard Fikes,et al.  Learning and Executing Generalized Robot Plans , 1993, Artif. Intell..

[7]  H. Simon,et al.  Perception in chess , 1973 .

[8]  S. Vera,et al.  Induction of Concepts in the Predicate Calculus , 1975, IJCAI.

[9]  Allen Newell,et al.  Computer science as empirical inquiry: symbols and search , 1976, CACM.

[10]  Gerald J. Sussman,et al.  Forward Reasoning and Dependency-Directed Backtracking in a System for Computer-Aided Circuit Analysis , 1976, Artif. Intell..

[11]  Tom Michael Mitchell,et al.  Model-directed learning of production rules , 1977, SGAR.

[12]  Ryszard S. Michalski,et al.  Selection of Most Representative Training Examples and Incremental Generation of VL1 Hypotheses: The Underlying Methodology and the Description of Programs ESEL and AQ11 , 1978 .

[13]  Frederick Hayes-Roth,et al.  An interference matching technique for inducing abstractions , 1978, CACM.

[14]  Tom Michael Mitchell.  Version spaces: an approach to concept learning. , 1979 .

[15]  Joshua Lederberg,et al.  Applications of Artificial Intelligence for Organic Chemistry: The DENDRAL Project , 1980 .

[16]  Michael R. Genesereth,et al.  Metaphors and Models , 1980, AAAI.

[17]  John McCarthy,et al.  Circumscription - A Form of Non-Monotonic Reasoning , 1980, Artif. Intell..

[18]  Gerald Jay Sussman,et al.  CONSTRAINTS - A Language for Expressing Almost-Hierarchical Descriptions , 1980, Artif. Intell..

[19]  Steve Tappel,et al.  Some Algorithm Design Methods , 1980, AAAI.

[20]  Thomas G. Dietterich,et al.  Inductive Learning of Structural Descriptions: Evaluation Criteria and Comparative Review of Selected Methods , 1981, Artif. Intell..

[21]  Tom M. Mitchell,et al.  Acquisition of Appropriate Bias for Inductive Concept Learning , 1982, AAAI.

[22]  Tom M. Mitchell,et al.  Generalization as Search , 2002 .

[23]  Allen Newell,et al.  The Knowledge Level , 1989, Artif. Intell..

[24]  Thomas G. Dietterich,et al.  Learning and Inductive Inference , 1982 .

[25]  Herbert A. Simon,et al.  WHY SHOULD MACHINES LEARN , 1983 .

[26]  Kurt Van Lehn,et al.  Felicity conditions for human skill acquisition: validating an al-based theory , 1983 .

[27]  James Arthur Gosling,et al.  Algebraic constraints , 1983 .

[28]  Douglas B. Lenat,et al.  Why AM and EURISKO Appear to Work , 1984, Artif. Intell..

[29]  Russell Greiner,et al.  What's New? A Semantic Definition of Novelty , 1983, IJCAI.

[30]  Bruce W. Porter,et al.  Episodic Learning , 1983, AAAI.

[31]  David J. Mostow,et al.  Machine Transformation of Advice Into a Heuristic Search Procedure , 1983 .

[32]  Richard M. Keller,et al.  Learning by Re-Expressing Concepts for Efficient Recognition , 1983, AAAI.

[33]  Pat Langley,et al.  Learning Effective Search Heuristics , 1983, IJCAI.

[34]  J. Ross Quinlan,et al.  Learning Efficient Classification Procedures and Their Application to Chess End Games , 1983 .

[35]  Paul D. Scott,et al.  Learning: The Construction of A Posteriori Knowledge Structures , 1983, AAAI.

[36]  Tom M. Mitchell,et al.  Learning and Problem Solving , 1983, IJCAI.

[37]  Agustin A. Araya,et al.  Learning Problem Classes by Means of Experimentation and Generalization , 1984, AAAI.

[38]  John McCarthy,et al.  Applications of Circumscription to Formalizing Common Sense Knowledge , 1987, NMR.

[39]  Joseph Y. Halpern,et al.  A Guide to the Modal Logics of Knowledge and Belief: Preliminary Draft , 1985, IJCAI.

[40]  Stuart J. Russell,et al.  The compleat guide to MRS , 1985 .

[41]  Sridhar Mahadevan,et al.  Verification-based learning: a generalisation strategy for inferring problem-reduction methods , 1985, IJCAI 1985.

[42]  Gerald DeJong,et al.  Learning Schemata for Natural Language Processing , 1985, IJCAI.

[43]  Sridhar Mahadevan,et al.  Verification-based Learning: A Generalized Strategy for Inferring Problem-Reduction Methods , 1985, IJCAI.

[44]  Thomas Glen Dietterich.  Constraint propagation techniques for theory-driven data interpretation (artificial intelligence, machine learning) , 1985 .

[45]  Thomas Ellman,et al.  Generalizing Logic Circuit Designs by Analyzing Proofs of Correctness , 1985, IJCAI.

[46]  Thomas G. Dietterich,et al.  Selecting Appropriate Representations for Learning from Examples , 1986, AAAI.

[47]  Thomas G. Dietterich,et al.  Exploiting functional vocabularies to learn structural descriptions , 1986 .

[48]  Thomas G. Dietterich.  The Test Incorporation Hypothesis and the Weak Methods , 1986 .

[49]  John E. Laird,et al.  Current research on learning in Soar , 1986 .

[50]  Tom M. Mitchell,et al.  Learning by experimentation: acquiring and refining problem-solving heuristics , 1993 .