Learning by Switching Type of Information

Authors: Sanjay Jain and Frank Stephan.

Source: Lecture Notes in Artificial Intelligence Vol. 2225, 2001, 205 - 218.

Abstract. The present work is dedicated to the study of modes of data-presentation between text and informant within the framework of inductive inference. The model is such that the learner requests sequences of positive and negative data and the relations between the various formalizations in dependence on the number of switches between positive and negative data is investigated. In particular it is shown that there is a proper hierarchy of the notions of learning from standard text, in the basic switching model, in the newtext switching model and in the restart switching model. The last one of these turns out to be equivalent to the standard notion of learning from informant.

©Copyright 2001 Springer