论文信息 - Transductive support vector machines for structured variables

Transductive support vector machines for structured variables

We study the problem of learning kernel machines transductively for structured output variables. Transductive learning can be reduced to combinatorial optimization problems over all possible labelings of the unlabeled data. In order to scale transductive learning to structured variables, we transform the corresponding non-convex, combinatorial, constrained optimization problems into continuous, unconstrained optimization problems. The discrete optimization parameters are eliminated and the resulting differentiable problems can be optimized efficiently. We study the effectiveness of the generalized TSVM on multiclass classification and label-sequence learning problems empirically.

[1] Thorsten Joachims,et al. Transductive Inference for Text Classification using Support Vector Machines , 1999, ICML.

[2] Ben Taskar,et al. Max-Margin Markov Networks , 2003, NIPS.

[3] Thomas Hofmann,et al. Hidden Markov Support Vector Machines , 2003, ICML.

[4] Xiaojin Zhu,et al. Kernel conditional random fields: representation and clique selection , 2004, ICML.

[5] Thomas Hofmann,et al. Support vector machine learning for interdependent and structured output spaces , 2004, ICML.

[6] Alexander Zien,et al. Semi-Supervised Classification by Low Density Separation , 2005, AISTATS.

[7] Mikhail Belkin,et al. Beyond the point cloud: from transductive to semi-supervised learning , 2005, ICML.

[8] Mikhail Belkin,et al. Maximum Margin Semi-Supervised Learning for Structured Variables , 2005, NIPS 2005.

[9] Alexander Zien,et al. Semi-Supervised Learning , 2006 .

[10] Ulf Brefeld,et al. Semi-supervised learning for structured output variables , 2006, ICML.

[11] Bernhard Schölkopf,et al. Introduction to Semi-Supervised Learning , 2006, Semi-Supervised Learning.

[12] Dale Schuurmans,et al. Discriminative unsupervised learning of structured predictors , 2006, ICML.

[13] Olivier Chapelle,et al. Training a Support Vector Machine in the Primal , 2007, Neural Computation.

[14] Thomas Hofmann,et al. Learning to Model Spatial Dependency: Semi-Supervised Discriminative Random Fields , 2007 .