Adaptive Regularization for Transductive Support Vector Machine
作者： Irwin King , Jianke Zhu , Rong Jin , Michael Lyu , Zenglin Xu
DOI:
关键词:
摘要: We discuss the framework of Transductive Support Vector Machine (TSVM) from perspective regularization strength induced by unlabeled data. In this framework, SVM and TSVM can be regarded as a learning machine without one with full data, respectively. Therefore, to supplement strength, it is necessary introduce data-dependant partial regularization. To end, we reformulate into form controllable which includes special cases. Furthermore, method adaptive that data dependant based on smoothness assumption. Experiments set benchmark sets indicate promising results proposed work compared state-of-the-art algorithms.
nips.cc参考文章(24)
R. Cowell Z. Ghahramani, A Zien, O Chapelle, Semi-Supervised Classification by Low Density Separation international conference on artificial intelligence and statistics. pp. 57- 64 ,(2005)
Dale Schuurmans, Linli Xu, Unsupervised and semi-supervised multi-class support vector machines national conference on artificial intelligence. pp. 904- 910 ,(2005)
Aarti Singh, Xiaojin Zhu, Robert Nowak, ERRATA for NIPS 2008 paper "Unlabeled data: Now it helps, now it doesn't" ,(2008)
Semi-Supervised Learning Advanced Methods in Sequence Analysis Lectures. pp. 221- 232 ,(2010) , 10.7551/MITPRESS/9780262033589.001.0001
Avrim Blum, Shuchi Chawla, Learning from Labeled and Unlabeled Data using Graph Mincuts international conference on machine learning. pp. 19- 26 ,(2001) , 10.1184/R1/6606860.V1
Steven C. H. Hoi, Rong Jin, Michael R. Lyu, Learning nonparametric kernel matrices from pairwise constraints international conference on machine learning. pp. 361- 368 ,(2007) , 10.1145/1273496.1273542
Steven C. H. Hoi, Michael R. Lyu, Edward Y. Chang, Learning the unified kernel machines for classification knowledge discovery and data mining. pp. 187- 196 ,(2006) , 10.1145/1150402.1150426
Vikas Sindhwani, Partha Niyogi, Mikhail Belkin, Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples Journal of Machine Learning Research. ,vol. 7, pp. 2399- 2434 ,(2006)
Partha Niyogi, Hariharan Narayanan, Mikhail Belkin, On the Relation Between Low Density Separation, Spectral Clustering and Graph Cuts neural information processing systems. pp. 1025- 1032 ,(2006)
Irwin King, Jianke Zhu, Rong Jin, Michael Lyu, Zenglin Xu, Efficient Convex Relaxation for Transductive Support Vector Machine neural information processing systems. ,vol. 20, pp. 1641- 1648 ,(2007)