Learning as Understanding the External World

In this paper we propose an abstract model of a learning system that represents a suitable framework for multistrategy task-adaptive. learning, a type of learning that integrates dynamically a whole range of learning strategies, depending of the features of the current learning task. The system has an incomplete and partially incorrect representation of the external world, in the knowledge base KB, and receives new input information from the environment. The learning goal is to extend, update and/or improve the KB, so as to consistently integrate the information contained in the input. This means that, after learning from an input I, the KB should be such that a generalization of I is inferable from it. The learning method is based on the idea of "understanding" the input through an exploration of the KB, and an employment of different inference types.

[1]  E. Feigenbaum,et al.  Computers and Thought , 1963 .

[2]  P. Winston.  Learning by Augmenting Rules and Accumulating Censors. , 1982 .

[3]  Dedre Gentner,et al.  Structure-Mapping: A Theoretical Framework for Analogy , 1983, Cogn. Sci..

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

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

[6]  Scott Bennett,et al.  A Domain Independent Explanation-Based Generalizer , 1986, AAAI.

[7]  Stuart J. Russell,et al.  A Logical Approach to Reasoning by Analogy , 1987, IJCAI.

[8]  Gerald DeJong,et al.  The Classification, Detection and Handling of Imperfect Theory Problems , 1987, IJCAI.

[9]  Jaime G. Carbonell,et al.  Strategies for Learning Search Control Rules: An Explanation-based Approach , 1987, IJCAI.

[10]  Andrea Pohoreckyj Danyluk,et al.  The Use of Explanations for Similarity-based Learning , 1987, IJCAI.

[11]  Michael J. Pazzani,et al.  Integrating Explanation-Based and Empirical Learning Methods in OCCAM , 1988, EWSL.

[12]  G. Tecuci.  Disciple : a theory, methodology and system for learning expert knowledge , 1988 .

[13]  Jude Shavlik,et al.  An Approach to Combining Explanation-based and Neural Learning Algorithms , 1989 .

[14]  Ryszard S. Michalski,et al.  The Logic of Plausible Reasoning: A Core Theory , 1989, Cogn. Sci..

[15]  Ray Bareiss,et al.  Concept Learning and Heuristic Classification in WeakTtheory Domains , 1990, Artif. Intell..

[16]  Francesco Bergadano,et al.  Guiding induction with domain theories , 1990 .

[17]  D. C. Wilkins.  Knowledge base refinement as improving an incorrect and incomplete domain theory , 1993 .

[18]  Stan Matwin,et al.  Explanation-Based Learning with Incomplete Theories: A Three-step Approach , 1990, ML.

[19]  Yves Kodratoff,et al.  Using Abductive Recovery of Failed Proofs for Problem Solving by Analogy , 1990, ML.

[20]  Ryszard S. Michalski,et al.  Inferential Learning Theory as a Basis for Multistrategy Task-Adaptive Learning , 1991 .

[21]  Gheorghe Tecuci.  Steps Toward Automating Knowledge Acquisition for Expert Systems , 1991 .

[22]  Gheorghe Tecuci,et al.  A Method for Multistrategy Task-Adaptive Learning Based on Plausible Justifications , 1991, ML.

[23]  Jaime G. Carbonell,et al.  Derivational analogy: a theory of reconstructive problem solving and expertise acquisition , 1993 .

[24]  Ryszard S. Michalski,et al.  A theory and methodology of inductive learning , 1993 .