Decremental multi-output least square SVR learning

Author

Zhang, Wei ; Liu, Xianhui ; Shi, Deming ; Wang, Weidong

Author_Institution

Eng. Res. Center for Enterprise Digital Technol. of Minist. of Educ., Tongji Univ., Shanghai, China

Volume

1

fYear

2011

fDate

10-12 June 2011

Firstpage

636

Lastpage

639

Abstract

The solution of multi-output LS-SVR machines follows from solving a set of linear equations. Compared with ε-intensive SVR, it loses the advantage of a sparse decomposition. In order to limit the number of support vectors and reduce the computation cost, this paper presents a decremental recursive algorithm for multi-output LS-SVR machines. This algorithm removes one sample one time and large-scale matrix inverse is computed quickly based on previous results. The decremental algorithm can be used to train online multi-output LS-SVR machine. Experimental results demonstrate the effectiveness of the algorithm.

Keywords

learning (artificial intelligence); least squares approximations; support vector machines; decremental algorithm; decremental recursive algorithm; large-scale matrix inverse; linear equations; multioutput least square SVR learning; online multioutput LS-SVR machine training; sparse decomposition; support vectors; Accuracy; Algorithm design and analysis; Equations; Mathematical model; Prediction algorithms; Support vector machines; Training; decremental recursive algorithm; ls-svr; matrix inverse; multi-output; sparse;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science and Automation Engineering (CSAE), 2011 IEEE International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-8727-1

Type

conf

DOI

10.1109/CSAE.2011.5953299

Filename

5953299