Explanation-based generalization: A unifying view

Abstract

Abstract. The problem of formulating general concepts from specific training examples has long been a major focus of machine learning research. While most previous research has focused on empirical methods for generalizing from a large number of training examples using no domain-specific knowledge, in the past few years new methods have been developed for applying domain-specific knowledge to formulate valid generalizations from single training examples. The characteristic common to these methods is that their ability to generalize from a single example follows from their ability to explain why the training example is a member of the concept being learned. This paper proposes a general, domain-independent mechanism, called EBG, that unifies previous approaches to explanation-based generalization. The EBG method is illustrated in the context of several example problems, and used to contrast several existing systems for explanation-based generalization. The perspective on explanation-based generalization afforded by this general method is also used to identify open research problems in this area.

Explanation-based learning: a survey of programs and perspectives

“…The EBG formalism is also useful for clarifying the relation between generaliza-"' The version of MGR in Mitchell et al [1986] contains an error that was pointed out and corrected in DeJong and . tion, chunking, operationalization, and analogy.…”

“…A formalism called explanation-based generalization (EBG)" has been proposed by Mitchell et al [1986]. EBG attempts to capture the essential elements of most explanation-based learning systems that have been proposed.…”

“…(Mitchell and Utgoff) Another major effort to investigate EBL techniques was undertaken by Mitchell and co-workers at Rutgers University. A number of different EBL systems were developed by Mitchell's group [Kedar-Cabeili 1985;Keller 1983;Mitchell 198213;Mitchell 1983;Mitchell et al 1986;Steinberg and Mitchell 1985;Utgoff 19861. One of the oldest of these systems is LEX-II, which learns search control heuristics in the domain of symbolic integration.…”

Rationality and Its Roles in Reasoning

“…For example, while DeJong's (1983) original criteria for explanation-based learning concerned utility of the result, many techniques of explanation-based generalization (Mitchell et al 1986) use only a categorical notion of operationality of concept definitions that avoids comparing the different explanations that might form the basis of learning. Similarly, many techniques have been developed for case-based reasoning, but few if any of these have been analyzed or interpreted in terms of the expected utility of the cases recorded or of the reconstruction process.…”

Combining explanation based generalization with the learning of macro operators (abstract only)

“…Explanation-based generalization (EBG) is a learning technique which attempts to learn operational concept definitions from an explanation or proof of why a particular example is a concept instance [1,2]. One way to create such a proof is through search of some proof problem space.…”