Semi-supervised multitask learning via self-training and maximum entropy discrimination

Authors:
Guoqing Chao;Shiliang Sun
Affiliations:
Department of Computer Science and Technology, East China Normal University, Shanghai, P.R. China;Department of Computer Science and Technology, East China Normal University, Shanghai, P.R. China
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part III
Year:
2012

Citing 8
Cited 0

Multitask Learning

Machine Learning - Special issue on inductive transfer
Semi-Supervised Self-Training of Object Detection Models

WACV-MOTION '05 Proceedings of the Seventh IEEE Workshops on Application of Computer Vision (WACV/MOTION'05) - Volume 1 - Volume 01
Effective self-training for parsing

HLT-NAACL '06 Proceedings of the main conference on Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics
Multi-Task Learning for Classification with Dirichlet Process Priors

The Journal of Machine Learning Research
Multiple-view multiple-learner active learning

Pattern Recognition
Active learning with extremely sparse labeled examples

Neurocomputing
Multitask Sparsity via Maximum Entropy Discrimination

The Journal of Machine Learning Research
Multi-view laplacian support vector machines

ADMA'11 Proceedings of the 7th international conference on Advanced Data Mining and Applications - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Maximum entropy discrimination (MED) is already shown to be effective for discriminative classification and regression, and can be applied to multitask learning (MTL) with some further assumptions. Self-training is a commonly used technique for semi-supervised learning. In order to integrate the merits offered by semi-supervised learning and MTL, this paper presents semi-supervised MTL via self-training and MED. We select the suitable measure metric and identify how to use unlabeled data. Experimental results on two UCI data sets demonstrate that our method yields better performance than semi-supervised single-task learning (STL) and supervised MTL.