conftrace_

2006 JMLR JMLR 2006

Large Scale Transductive SVMs

Abstract

We show how the concave-convex procedure can be applied to transductive SVMs, which traditionally require solving a combinatorial search problem. This provides for the first time a highly scalable algorithm in the nonlinear case. Detailed experiments verify the utility of our approach. Software is available at http://www.kyb.tuebingen.mpg.de/bs/people/fabee/transduction.html. [abs] [ pdf ][ bib ] © JMLR 2006. (edit, beta)

🐣 Hot Topic Early Bird - semi-supervised learning

Authors

Ronan Collobert, Fabian Sinz, Jason Weston, Léon Bottou

Topics

Machine Learning > Core Methods > Classification Machine Learning > Learning Types > Semi-Supervised Learning Machine Learning > Core Methods > Support Vector Machine

Keywords

semi-supervised learning transductive learning concave-convex procedure support vector machine scalable algorithm

More from JMLR 2006

All JMLR 2006 papers →

Kernel-Based Learning of Hierarchical Multilabel Classification Models

Structured Prediction, Dual Extragradient and Bregman Projections

Active Learning with Feedback on Features and Instances

Considering Cost Asymmetry in Learning Classifiers

Learning Factor Graphs in Polynomial Time and Sample Complexity

Collaborative Multiagent Reinforcement Learning by Payoff Propagation