A Linear Programming Neural Circuit Model
In this paper we present a neural circuit model which can solve linear programming problems. The main feature of the model is that it takes into account saturating behaviour of circuit elements in a possible realizations. The neural model can be viewed as a gradient system and its operation is based on the penalty function approach of solving linear programming tasks. In the paper the properties of the model are discussed including stability and that how to utilize the saturation in the neurons for obtaining better performance.
KeywordsNeural Network Penalty Function Linear Programming Problem Recurrent Neural Network Saturation Region
Unable to display preview. Download preview PDF.
- J. Bíró, E. Halasz, T. Trón, M. Boda, and G. Privitzky. Neural networks for exact constrained optimization. In Lecture Notes in Computer Science (Proceedings of the International Conference on Artificial Neural Networks-ICANN’96), Bochum (Germany), June 1996.Google Scholar
- J. Bíró, Z. Koronkai, L. Ast, T. Trón, and M. Boda. Analyses of extended and generalized optimization neural networks. Journal of Artificial Neural Networks, 2(4):401–409, August 1995.Google Scholar
- L.O. Chua and G. Lin. Nonlinear programming without computation. IEEE Trans, on Circuits and Systems, 52(2), February 1984.Google Scholar
- K. Zikan T.P. Caudell. A neural network architecture for linear programming. In Proc. of IEEE ICNN’92, pages 91–96, October 1992.Google Scholar