next up previous
Next: Introduction

Active Learning with Statistical Models

David A. Cohn
Zoubin Ghahramani
Michael I. Jordan

Center for Biological and Computational Learning
Dept. of Brain and Cognitive Sciences
Massachusetts Institute of Technology
Cambridge, MA 02139 USA


For many types of machine learning algorithms, one can compute the statistically ``optimal'' way to select training data. In this paper, we review how optimal data selection techniques have been used with feedforward neural networks. We then show how the same principles may be used to select data for two alternative, statistically-based learning architectures: mixtures of Gaussians and locally weighted regression. While the techniques for neural networks are computationally expensive and approximate, the techniques for mixtures of Gaussians and locally weighted regression are both efficient and accurate. Empirically, we observe that the optimality criterion sharply decreases the number of training examples the learner needs in order to achieve good performance.

David Cohn
Mon Mar 25 09:20:31 EST 1996