Not Logged In

Unsupervised and Semi-Supervised Multi-Class Support Vector Machines

We present new unsupervised and semi-supervised training algorithms for multi-class support vector machines based on semidefinite programming. Although support vector machines (SVMs) have been a dominant machine learning technique for the past decade, they have generally been applied to supervised learning problems. Developing unsupervised extensions to SVMs has in fact proved to be difficult. In this paper, we present a principled approach to unsupervised SVM training by formulating convex relaxations of the natural training criterion: find a labeling that would yield an optimal SVM classifier on the resulting training data. The problem is hard, but semidefinite relaxations can approximate this objective surprisingly well. While previous work has concentrated on the two-class case, we present a general, multi-class formulation that can be applied to a wider range of natural data sets. The resulting training procedures are computationally intensive, but produce high quality generalization results.

Citation

L. Xu, D. Schuurmans. "Unsupervised and Semi-Supervised Multi-Class Support Vector Machines". National Conference on Artificial Intelligence (AAAI), Pittsburgh, January 2005.

Keywords: vector, multi-class, machine learning
Category: In Conference

BibTeX

@incollection{Xu+Schuurmans:AAAI05,
  author = {Linli Xu and Dale Schuurmans},
  title = {Unsupervised and Semi-Supervised Multi-Class Support Vector
    Machines},
  booktitle = {National Conference on Artificial Intelligence (AAAI)},
  year = 2005,
}

Last Updated: April 25, 2007
Submitted by William Thorne

University of Alberta Logo AICML Logo