Metric and kernel learning arise in several machine learning applications. However, most existing metric learning algorithms are limited to learning metrics over low-dimensional d...
Prateek Jain, Brian Kulis, Jason V. Davis, Inderji...
Abstract. We propose a new representation for the domain of Definite Boolean functions. The key idea is to view the set of models of a Boolean function as an incidence relation be...
We explore an application to the game of Go of a reinforcement learning approach based on a linear evaluation function and large numbers of binary features. This strategy has prov...
We define the base polytope B(P, g) of a partially ordered set P and a supermodular function g on the ideals ofP as the convex hull of the incidence vectors of all linear extensio...
—Discriminative Training (DT) methods for acoustic modeling, such as MMI, MCE, and SVM, have been proved effective in speaker recognition. In this paper we propose a DT method fo...