Multitask multiclass privileged information support vector machines
作者： Shiliang Sun , You Ji , Yue Lu
DOI:
关键词:
摘要: In this paper, we propose a new learning paradigm named multitask multiclass privileged information support vector machines. The starting point of our work is mainly based on the success machines which cast problems as constrained optimization problem with quadratic objective function. Learning using an advanced integrated idea human teaching in machine learning. This paper extends multi-class to strategy. Our approach can take full advantages and information. Experimental results show that approaches obtains very good for problems.
参考文章(8)
Dmitry Pechyony, Akshay Vashist, Vladimir Vapnik, Rauf Izmailov, SMO-Style Algorithms for Learning Using Privileged Information. DMIN. pp. 235- 241 ,(2010)
You Ji, Shiliang Sun, Multitask Multiclass Support Vector Machines international conference on data mining. pp. 512- 518 ,(2011) , 10.1109/ICDMW.2011.126
Kilian Q. Weinberger, Shibin Parameswaran, Large Margin Multi-Task Metric Learning neural information processing systems. ,vol. 23, pp. 1867- 1875 ,(2010)
Theodoros Evgeniou, Massimiliano Pontil, Regularized multi--task learning knowledge discovery and data mining. pp. 109- 117 ,(2004) , 10.1145/1014052.1014067
Theodoros Evgeniou, Charles A Micchelli, Massimiliano Pontil, John Shawe-Taylor, Learning Multiple Tasks with Kernel Methods Journal of Machine Learning Research. ,vol. 6, pp. 615- 637 ,(2005) , 10.5555/1046920.1088693
Shiliang Sun, Multitask learning for EEG-based biometrics international conference on pattern recognition. pp. 1- 4 ,(2008) , 10.1109/ICPR.2008.4761865
Koby Crammer, Yoram Singer, On the algorithmic implementation of multiclass kernel-based vector machines Journal of Machine Learning Research. ,vol. 2, pp. 265- 292 ,(2002)
Rich Caruana, Multitask learning Learning to learn. pp. 95- 133 ,(1998)