Discrimination net models of concept formation

Publisher Summary This chapter discusses some of the earliest discrimination net systems, which have traditionally been forwarded as models of recognition and rote learning. There are two particular systems that use discrimination nets for concept learning, one that is supervised and non-incremental, and another that is unsupervised and incremental. The unsupervised and incremental model is known as EPAM. EPAM is an early learning system that has greatly influenced the line of research known as concept formation. The recent versions of EPAM explain certain perceptual and learning phenomena, notably context effects in letter perception, as well as more universal typicality phenomena. Together with its influence on more recent work, this suggests that discrimination net models continue to have something to offer as models of concept formation. The chapter also discusses the relationship between EPAM and other computational models of learning, including chunking systems that gradually combine pieces of conceptual information over a stream of data.

[1]  James L. McClelland,et al.  An interactive activation model of context effects in letter perception: Part 2. The contextual enhancement effect and some tests and extensions of the model. , 1982, Psychological review.

[2]  G. A. Miller.  THE PSYCHOLOGICAL REVIEW THE MAGICAL NUMBER SEVEN, PLUS OR MINUS TWO: SOME LIMITS ON OUR CAPACITY FOR PROCESSING INFORMATION 1 , 1956 .

[3]  H. Simon,et al.  STM capacity for Chinese words and idioms: Chunking and acoustical loop hypotheses , 1985, Memory & cognition.

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

[5]  James L. McClelland,et al.  An interactive activation model of context effects in letter perception: I. An account of basic findings. , 1981 .

[6]  John R. Anderson.  Retrieval of propositional information from long-term memory , 1974 .

[7]  Herbert A. Simon,et al.  Context effects in letter perception: comparison of two theories. , 1988, Psychological review.

[8]  E. Rosch.  ON THE INTERNAL STRUCTURE OF PERCEPTUAL AND SEMANTIC CATEGORIES1 , 1973 .

[9]  H. Simon,et al.  A simulation of memory for chess positions. , 1973 .

[10]  E. R. Crossman.  A THEORY OF THE ACQUISITION OF SPEED-SKILL∗ , 1959 .

[11]  C. E. SHANNON,et al.  A mathematical theory of communication , 1948, MOCO.

[12]  R. Seibel.  DISCRIMINATION REACTION TIME FOR 1,023-ALTERNATIVE TASK. , 1963, Journal of experimental psychology.

[13]  E. Rosch,et al.  Family resemblances: Studies in the internal structure of categories , 1975, Cognitive Psychology.

[14]  H. Simon,et al.  An information-processing theory of some effects of similarity, familiarization, and meaningfulness in verbal learning , 1964 .

[15]  Janet L. Kolodner,et al.  Reconstructive Memory: A Computer Model , 1983, Cogn. Sci..

[16]  James C Johnston,et al.  A test of the Sophisticated Guessing Theory of word perception , 1978, Cognitive Psychology.

[17]  U. Neisser,et al.  Hierarchies in concept attainment. , 1962, Journal of experimental psychology.

[18]  H. Simon,et al.  A theory of the serial position effect. , 1962, British journal of psychology.

[19]  Herbert A. Simon,et al.  An information-processing explanation of one-trial and incremental learning , 1967 .

[20]  H. Simon.  Amounts of fixation and discovery in maze learning behavior , 1957 .

[21]  Laurent Siklossy,et al.  Natural language learning by computer , 1968 .

[22]  Douglas H. Fisher,et al.  The Structure and Formation of Natural Categories , 1990 .

[23]  H. Simon,et al.  EPAM-like Models of Recognition and Learning , 1984, Cogn. Sci..

[24]  Lance J. Rips,et al.  Semantic distance and the verification of semantic relations , 1973 .

[25]  J. Bruner,et al.  Familiarity of Letter Sequences and Tachistoscopic Identification , 1954 .

[26]  Douglas L. Hintzman,et al.  Explorations with a discrimination net model for paired-associate learning , 1968 .