Abstract
Online semi-supervised learning (OS2L) has received much attention recently because of its well practical usefulness. Most of existing studies of online semi-supervised learning related to manifold regularization. In this study, we propose a novel online semi-supervised learning framework with multiple regularization terms based on the notion of ascending the dual function in constrained optimization. Using Fenchel conjugate, different semi-supervised regularization terms can be integrated in the dual function easily and directly. This approach is derived by updating limited dual coefficient variables on each learning round. To be practical, we also employ buffering strategies and sparse approximation approaches in this work. Experimental studies show that our methods achieve accuracy comparable to offline algorithms while consuming less time and memory. Specially, our online semi-supervised learning algorithms can handle the settings where the target hyperplane of classification continually drifts with the sequence of arriving instances. This paper paves a way to the design and analysis of online semi-supervised learning algorithms with multiple regularization terms.
Funding Information
  • National Natural Science Foundation of China (71471174)

This publication has 31 references indexed in Scilit: