论文信息 - Variations on a Kernel-Adatron Theme

Variations on a Kernel-Adatron Theme

The Kernel-Adatron (KA) algorithm was recently introduced as an alternative to quadratic programming for training support vector machines. In this paper we investigate three variants of the original KA algorithm and demonstrate through examples that they deliver excellent accuracy. The examples also show that one of these variants has rather different learning properties to the other two.

Tom Downs | Hugo D. Navone

[1] Terrence J. Sejnowski,et al. Analysis of hidden units in a layered network trained to classify sonar targets , 1988, Neural Networks.

[2] Nello Cristianini,et al. Simple Learning Algorithms for Training Support Vector Machines , 1998 .

[3] Yoav Freund,et al. Large Margin Classification Using the Perceptron Algorithm , 1998, COLT' 98.

[4] Nello Cristianini,et al. The Kernel-Adatron Algorithm: A Fast and Simple Learning Procedure for Support Vector Machines , 1998, ICML.

[5] Yoav Freund,et al. Large Margin Classification Using the Perceptron Algorithm , 1998, COLT.

[6] Federico Girosi,et al. Support Vector Machines: Training and Applications , 1997 .

[7] Vladimir N. Vapnik,et al. The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[8] W. Kinzel. Statistical mechanics of the perceptron with maximal stability , 1990 .

[9] Lutgarde M. C. Buydens,et al. Using support vector machines for time series prediction , 2003 .

[10] M. Aizerman,et al. Theoretical Foundations of the Potential Function Method in Pattern Recognition Learning , 1964 .