A Co-Regularization Approach to Semi-supervised Learning with Multiple Views

Abstract

The Co-Training algorithm uses unlabeled examples in multiple views to bootstrap classifiers in each view, typically in a greedy manner, and operating under assumptions of view-independence and compatibility. In this paper, we propose a Co-Regularization framework where classifiers are learnt in each view through forms of multi-view regularization. We propose algorithms within this framework that are based on optimizing measures of agreement and smoothness over labeled and unlabeled examples. These algorithms naturally extend standard regularization methods like Support Vector Machines (SVM) and Regularized Least squares (RLS) for multi-view semi-supervised learning, and inherit their benefits and applicability to high-dimensional classification problems. An empirical investigation is presented that confirms the promise of this approach.