Efficient large margin semisupervised learning

Junhui Wang
; Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, PMLR 2:588-595, 2007.

Abstract

In classification, semisupervised learning involves a large amount of unlabeled data with only a small number of labeled data. This imposes great challenge in that the class probability given input can not be well estimated through labeled data alone. To enhance predictability of classification, this article introduces a large margin semisupervised learning method constructing an efficient loss to measure the contribution of unlabeled instances to classification. The loss is iteratively refined, based on which an iterative scheme is derived for implementation. The proposed method is examined for two large margin classifiers: support vector machines and ψ-learning. Our theoretical and numerical analyses indicate that the method achieves the desired objective of delivering higher performances over any other method initializing the scheme.

Cite this Paper


BibTeX
@InProceedings{pmlr-v2-wang07b, title = {Efficient large margin semisupervised learning}, author = {Junhui Wang}, booktitle = {Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics}, pages = {588--595}, year = {2007}, editor = {Marina Meila and Xiaotong Shen}, volume = {2}, series = {Proceedings of Machine Learning Research}, address = {San Juan, Puerto Rico}, month = {21--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v2/wang07b/wang07b.pdf}, url = {http://proceedings.mlr.press/v2/wang07b.html}, abstract = {In classification, semisupervised learning involves a large amount of unlabeled data with only a small number of labeled data. This imposes great challenge in that the class probability given input can not be well estimated through labeled data alone. To enhance predictability of classification, this article introduces a large margin semisupervised learning method constructing an efficient loss to measure the contribution of unlabeled instances to classification. The loss is iteratively refined, based on which an iterative scheme is derived for implementation. The proposed method is examined for two large margin classifiers: support vector machines and ψ-learning. Our theoretical and numerical analyses indicate that the method achieves the desired objective of delivering higher performances over any other method initializing the scheme.} }
Endnote
%0 Conference Paper %T Efficient large margin semisupervised learning %A Junhui Wang %B Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2007 %E Marina Meila %E Xiaotong Shen %F pmlr-v2-wang07b %I PMLR %J Proceedings of Machine Learning Research %P 588--595 %U http://proceedings.mlr.press %V 2 %W PMLR %X In classification, semisupervised learning involves a large amount of unlabeled data with only a small number of labeled data. This imposes great challenge in that the class probability given input can not be well estimated through labeled data alone. To enhance predictability of classification, this article introduces a large margin semisupervised learning method constructing an efficient loss to measure the contribution of unlabeled instances to classification. The loss is iteratively refined, based on which an iterative scheme is derived for implementation. The proposed method is examined for two large margin classifiers: support vector machines and ψ-learning. Our theoretical and numerical analyses indicate that the method achieves the desired objective of delivering higher performances over any other method initializing the scheme.
RIS
TY - CPAPER TI - Efficient large margin semisupervised learning AU - Junhui Wang BT - Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics PY - 2007/03/11 DA - 2007/03/11 ED - Marina Meila ED - Xiaotong Shen ID - pmlr-v2-wang07b PB - PMLR SP - 588 DP - PMLR EP - 595 L1 - http://proceedings.mlr.press/v2/wang07b/wang07b.pdf UR - http://proceedings.mlr.press/v2/wang07b.html AB - In classification, semisupervised learning involves a large amount of unlabeled data with only a small number of labeled data. This imposes great challenge in that the class probability given input can not be well estimated through labeled data alone. To enhance predictability of classification, this article introduces a large margin semisupervised learning method constructing an efficient loss to measure the contribution of unlabeled instances to classification. The loss is iteratively refined, based on which an iterative scheme is derived for implementation. The proposed method is examined for two large margin classifiers: support vector machines and ψ-learning. Our theoretical and numerical analyses indicate that the method achieves the desired objective of delivering higher performances over any other method initializing the scheme. ER -
APA
Wang, J.. (2007). Efficient large margin semisupervised learning. Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, in PMLR 2:588-595

Related Material