GA-MTL: A Random Method of Multi-Task Learning

Author

Liu, Tsung-Yen ; Li, Guo-zheng ; Wu, G.-F. ; Chi, Eric C.

Author_Institution

School of Computer Engineering & Science, Shanghai University, Shanghai 200072, China, State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210093, China. Phone: +86-21-56335263, Fax: +86-21-56333061

fYear

2006

fDate

Oct. 2006

Firstpage

1762

Lastpage

1765

Abstract

Multi-task learning techniques can employ the removed redundant information to improve prediction accuracy. Which features to add to the target and/or the input during multi-task learning is still an open issue. The previous study used heuristic search methods. In this paper, a random method of genetic algorithm based multi-task learning (GA-MTL) is proposed to automatically determine the features for the input and/or the target. Experimental results on data sets from the real world show that GA-MTL is easy to use and obtains better performance than heuristic methods.

Keywords

Accuracy; Application software; Filters; Genetic algorithms; Laboratories; Learning systems; Machine learning; Neural networks; Search methods; Systems engineering and theory; Feature Selection; Genetic Algorithm; Multi-Task Learning; Neural Networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Engineering in Systems Applications, IMACS Multiconference on

Conference_Location

Beijing, China

Print_ISBN

7-302-13922-9

Electronic_ISBN

7-900718-14-1

Type

conf

DOI

10.1109/CESA.2006.313598

Filename

4105664