Part of Advances in Neural Information Processing Systems 15 (NIPS 2002)
We address the question of feature selection in the context of visual recognition. It is shown that, besides efﬁcient from a computational standpoint, the infomax principle is nearly optimal in the minimum Bayes error sense. The concept of marginal diversity is introduced, lead- ing to a generic principle for feature selection (the principle of maximum marginal diversity) of extreme computational simplicity. The relation- ships between infomax and the maximization of marginal diversity are identiﬁed, uncovering the existence of a family of classiﬁcation proce- dures for which near optimal (in the Bayes error sense) feature selection does not require combinatorial search. Examination of this family in light of recent studies on the statistics of natural images suggests that visual recognition problems are a subset of it.