A new neural network for solving linear programming problems

A. Cichocki, R. Unbehauen, K. Weinzierl, R. Hölzel

Research output: Contribution to journalArticlepeer-review

30 Citations (Scopus)

Abstract

We propose and analyse a new class of neural network models for solving linear programming (LP) problems in real time. We introduce a novel energy function that transforms linear programming into a system of nonlinear differential equations. This system of differential equations can be solved on-line by a simplified low-cost analog neural network containing only one single artificial neuron with adaptive synaptic weights. The network architecture is suitable for currently available CMOS VLSI implementations. An important feature of the proposed neural network architecture is its flexibility and universality. The correctness and performance of the proposed neural network is illustrated by extensive computer simulation experiments.

Original languageEnglish
Pages (from-to)244-256
Number of pages13
JournalEuropean Journal of Operational Research
Volume93
Issue number2
DOIs
Publication statusPublished - 6 Sep 1996
Externally publishedYes

Keywords

  • Linear programming
  • Neural networks
  • Parallel computing
  • Stochastic gradient descent optimization

Fingerprint

Dive into the research topics of 'A new neural network for solving linear programming problems'. Together they form a unique fingerprint.

Cite this