A Simpler Look at Consistency †
نویسندگان
چکیده
One of the major goals of most early concept learners was to find hypotheses that were perfectly consistent with the training data. It was believed that this goal would indirectly achieve a high degree of predictive accuracy on a set of test data. Later research has partially disproved this belief. However, the issue of consistency has not yet been resolved completely. We examine the issue of consistency from a new perspective. To avoid overfitting the training data, a considerable number of current systems have sacrificed the goal of learning hypotheses that are perfectly consistent with the training instances by setting a goal of hypothesis simplicity (Occam’s razor). Instead of using simplicity as a goal, we have developed a novel approach that addresses consistency directly. In other words, our concept learner has the explicit goal of selecting the most appropriate degree of consistency with the training data. We begin this paper by exploring concept learning with less than perfect consistency. Next, we describe a system that can adapt its degree of consistency in response to feedback about predictive accuracy on test data. Finally, we present the results of initial experiments that begin to address the question of how tightly hypotheses should fit the training data for different problems.
منابع مشابه
Exploring Culture-Related Content in English Textbooks: A Closer Look at Advanced Series of Iran Language Institute
The aim of this article was to examine three advanced textbooks in Iran Language Institute (ILI) in an attempt to establish if they differ in the extent to which they represent dimension of big ‘C’ culture and little ‘c’ culture, their stance in distribution of references of cultural category, and also what themes predominate. The analysis identifies just the cultural elements, and culture–free...
متن کاملConflict-Directed Backjumping Revisited
In recent years, many improvements to backtracking algorithms for solving constraint satisfaction problems have been proposed. The techniques for improving backtracking algorithms can be conveniently classiied as look-ahead schemes and look-back schemes. Unfortunately , look-ahead and look-back schemes are not entirely orthogonal as it has been observed empirically that the enhancement of look-...
متن کاملConnict-directed Backjumping Revisited
In recent years, many improvements to backtracking algorithms for solving constraint satisfaction problems have been proposed. The techniques for improving backtracking algorithms can be conveniently classi ed as look-ahead schemes and look-back schemes. Unfortunately, look-ahead and look-back schemes are not entirely orthogonal as it has been observed empirically that the enhancement of look-a...
متن کاملSimpler and Incremental Consistency Checking and Arc Consistency Filtering Algorithms for the Weighted Spanning Tree Constraint
The weighted spanning tree contraint is defined from a set of variables X and a value K. The variables X represent the nodes of a graph and the domain of a variable x ∈ X the neighbors of the node in the graph. In addition each pair (variable, value) is associated with a cost. This constraint states that the graph defined from the variables and the domains of the variables admits a spanning tre...
متن کامل