Mitchell, T. M.


Explanation-Based Generalization: A Unifying View

Classics

"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." Machine Learning, 1 (1), 47–80.


Explanation-Based Generalization: A Unifying View

Classics

"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." Machine Learning, 1 (1), 47–80.



Generalization as Search

Classics

Toward this end, generalization is cast as a search problem, and alternative methods for generalization are characterized in terms of the search strategies that they employ. Given the instance language, generalization language, and matching predicate, the generalization problem is to infer the identity of some unknown "target" generalization by observing a sample set of its training instances. Each training instance is an instance from the given language, along with its classification as either an instance of the target generalization (positive instance) or not an instance of the target generalization (negative instance). The generalization language corresponds to an hypothesis space (search space) of Possible solutions, and the learning task is to examine this hypothesis space, subject to constraints imposed by the training instances, to determine Plausible generalizations.