Coupled local minimizers: alternative formulations and extensions

Author

Suykens, J.A.K. ; Vandewalle, J.

Author_Institution

Dept. of Electr. Eng., Katholieke Univ., Leuven, Belgium

Volume

3

fYear

2002

fDate

6/24/1905 12:00:00 AM

Firstpage

2039

Lastpage

2043

Abstract

Coupled local minimizers (CLM) are a new method for cooperative searching which incorporates the principles of state synchronization and cellular neural networks (CNNs). Information between the optimizers is exchanged through coupling terms. The interaction between the individual minimizers can be optimized by solving linear programs. The original CLM method has been formulated within a Lagrange programming framework with equality constraints imposed between the individual optimizers. In this paper, we discuss alternative formulations and extensions to the original formulation

Keywords

cellular neural nets; constraint theory; cooperative systems; linear programming; minimisation; search problems; synchronisation; Lagrange programming framework; alternative formulations; cellular neural networks; cooperative searching; coupled local minimizers; coupling terms; equality constraints; linear programs; optimizer information exchange; state synchronization; Cellular neural networks; Constraint optimization; Cost function; Coupling circuits; Electronic mail; Image processing; Lagrangian functions; Master-slave; Optimization methods; Robots;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on

Conference_Location

Honolulu, HI

ISSN

1098-7576

Print_ISBN

0-7803-7278-6

Type

conf

DOI

10.1109/IJCNN.2002.1007453

Filename

1007453