A Multi-objective Learning Algorithm for RBF Neural Network

Author

Kokshenev, Illya ; Braga, Antonio Padua

Author_Institution

Depto. Eng. Eletron., Univ. Fed. de Minas Gerais, Belo Horizonte

fYear

2008

fDate

26-30 Oct. 2008

Firstpage

9

Lastpage

14

Abstract

In this paper, the problem of multi-objective supervised learning is discussed within the non-evolutionary optimization framework. The proposed MOBJ learning algorithm performs the search of Pareto-optimal models determining weights,width, prototype vectors, and the quantity of basis functions of the RBF network. In combination with the Akaike information criterion, the algorithm provides high quality solutions.

Keywords

Pareto optimisation; learning (artificial intelligence); radial basis function networks; search problems; Akaike information criterion; Pareto-optimal search; RBF neural network; multiobjective supervised learning algorithm; nonevolutionary optimization framework; Machine learning; Machine learning algorithms; Minimization methods; Neural networks; Optimization methods; Prototypes; Radial basis function networks; Risk management; Statistical learning; Supervised learning; LASSO; generalization; multi-objective learning; radial basis functions; regularization;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2008. SBRN '08. 10th Brazilian Symposium on

Conference_Location

Salvador

ISSN

1522-4899

Print_ISBN

978-1-4244-3219-6

Electronic_ISBN

1522-4899

Type

conf

DOI

10.1109/SBRN.2008.39

Filename

4665884