Semi-Supervised / Transductive Learning

Dear all,

This Wednesday (2:15-3:15 pm) I will be discussing the following paper:

@incollection{NIPS2006_357,
title = {Branch and Bound for Semi-Supervised Support Vector Machines},
author = {Olivier Chapelle and Vikas Sindhwani and Sathiya Keerthi},
booktitle = {Advances in Neural Information Processing Systems 19},
editor = {B. Sch\"{o}lkopf and J. Platt and T. Hoffman},
publisher = {MIT Press},
address = {Cambridge, MA},
pages = {},
year = {2007}
}

The abstract is:

Semi-supervised SVMs (S3 VM) attempt to learn low-density separators by maximizing the margin over labeled and unlabeled examples. The associated optimization problem is non-convex. To examine the full potential of S3 VMs modulo local minima problems in current implementations, we apply branch and bound techniques for obtaining exact, globally optimal solutions. Empirical evidence suggests that the globally optimal solution can return excellent generalization performance in situations where other implementations fail completely. While our current implementation is only applicable to small datasets, we discuss variants that can potentially lead to practically useful algorithms.

Thank You
Rashmin.