Learning to Model Spatial Dependency: Semi-Supervised Discriminative Random Fields
- Chi-Hoon Lee, Dept of Computing Science
- Shaojun Wang, Dept of Computing Science
- Feng Jiao, School of Computer Science, University of Waterloo
- Dale Schuurmans, AICML
- Russ Greiner, Dept of Computing Science; PI of AICML
We present a novel, semi-supervised approach to training discriminative random fields (DRFs) that efficiently exploits labeled and unlabeled training data to achieve improved accuracy in a variety of image processing tasks. We formulate DRF training as a form of MAP estimation that combines conditional loglikelihood on labeled data, given a data-dependent prior, with a conditional entropy regularizer defined on unlabeled data. Although the training objective is no longer concave, we develop an efficient local optimization procedure that produces classifiers that are more accurate than ones based on standard supervised DRF training. We then apply our semi-supervised approach to train DRFs to segment both synthetic and real data sets, and demonstrate significant improvements over supervised DRFs in each case.
Citation
C. Lee, S. Wang, F. Jiao, D. Schuurmans, R. Greiner. "Learning to Model Spatial Dependency: Semi-Supervised Discriminative Random Fields". Neural Information Processing Systems (NIPS), December 2006.| Keywords: | Semi-supervised learning, Random Fields, Structured outputs, machine learning, brain tumor, btap |
| Category: | In Conference |
BibTeX
@incollection{Lee+al:NIPS06,
author = {Chi-Hoon Lee and Shaojun Wang and Feng Jiao and Dale Schuurmans and
Russ Greiner},
title = {Learning to Model Spatial Dependency: Semi-Supervised Discriminative
Random Fields},
booktitle = {Neural Information Processing Systems (NIPS)},
year = 2006,
}Last Updated: February 11, 2013Submitted by Russ Greiner
