A NEW NEURAL-NETWORK FOR SOLVING LINEAR-PROGRAMMING PROBLEMS

We propose and analyse a new class of neural network models for solvin g linear programming (LP) problems in real time. We introduce a novel energy function that transforms linear programming into a system of no nlinear differential equations. This system of differential equations can be solved on-line by a simplified low-cost analog neural network c ontaining only one single artificial neuron with adaptive synaptic wei ghts, The network architecture is suitable for currently available CMO S VLSI implementations. An important feature of the proposed neural ne twork architecture is its flexibility and universality, The correctnes s and performance of the proposed neural network is illustrated by ext ensive computer simulation experiments.