Elsevier

Artificial Intelligence

Volume 20, Issue 2, February 1983, Pages 111-161
Artificial Intelligence

A theory and methodology of inductive learning

https://doi.org/10.1016/0004-3702(83)90016-4Get rights and content

Abstract

A theory of inductive learning is presented that characterizes it as a heuristic search through a space of symbolic descriptions, generated by an application of certain inference rules to the initial observational statements (the teacher-provided examples of some concepts, or facts about a class of objects or a phenomenon). The inference rules include generalization rules, which perform generalizing transformations on descriptions, and conventional truth-preserving deductive rules (specialization and reformulation rules). The application of the inference rules to descriptions is constrained by problem background knowledge, and guided by criteria evaluating the ‘quality’ of generated inductive assertions.

Based on this theory, a general methodology for learning structural descriptions from examples, called star, is described and illustrated by a problem from the area of conceptual data analysis.

References (90)

  • R.B. Banerji
  • A.W. Bierman et al.

    Survey of results in grammatical inference

  • A.W. Bierman

    The inference of regular LISP programs from examples

    IEEE Trans. Systems Man Cybernet.

    (1978)
  • M.M. Bongard
  • R.T. Brachman

    On the epistomological status of semantic networks

  • J.S. Bruner et al.
  • G.B. Buchanan et al.

    “Dendral and Meta-Dendral, their applications dimension

    Artificial Intelligence

    (1978)
  • B.G. Buchanan et al.

    Models of learning systems

  • R.M. Burstall et al.

    A transformation system for developing recursive programs

    J. ACM

    (1977)
  • R. Carnap

    The aim of inductive logic

  • J. Case et al.

    Comparison of identification criteria for mechanized inductive inference

  • C. Chang et al.
  • B.L. Cohen

    A powerful and efficient structural pattern recognition system

    Artificial Intelligence

    (1977)
  • R. Davis et al.
  • T. Dietterich

    Description of inductive program INDUCE 1.1

  • T. Dietterich

    A methodology of knowledge layers for inducing descriptions of sequentially ordered events

  • E.A. Feigenbaum

    The simulation of verbal learning behavior

  • B.R. Gaines

    Maryanski's grammatical inferences

    IEEE Trans. Comput.

    (1979)
  • J. Gaschnig

    Development of uranium exploration models for prospector consultant system

  • N. Hass et al.

    An approach to applying and acquiring knowledge

  • P. Hájek et al.

    The GUHA method of automatic hypothesis determination

    Computing

    (1966)
  • P. Hájek et al.
  • F. Hayes-Roth

    A structural approach to pattern learning and the acquisition of classificatory power

  • F. Hayes-Roth et al.

    An interference matching technique for inducing abstractions

    Comm. ACM

    (1978)
  • C.L. Hedrick

    A computer program to learn production systems using a semantic net

  • D.L. Hintzman
  • B. Hoff et al.

    INDUCE 2—a program for learning structural descriptions from examples

  • C.I. Hovland

    A ‘communication analysis’ of concept learning

    Psychol. Rev.

    (1952)
  • E.B. Hunt
  • E.B. Hunt et al.
  • J.P. Jouannaud et al.

    An automatic construction of LISP programs by transformations of functions synthesized from their input-output behavior

    Internat. J. Policy Anal. Inform. Systems

    (1980)
  • T.G. Kemeni

    The use of simplicity in induction

    Psychol. Rev.

    (1953)
  • M. Kochen

    Experimental study of hypothesis-formation by computer

  • P. Langley et al.

    A domain-independent framework for learning procedures

    Internat. J. Policy Anal. Inform. Systems

    (1980)
  • J. Larson

    Inductive inference in the variable-valued predicate logic system VL21: methodology and computer implementation

  • Cited by (0)

    View full text