Adjustment of parameters for signal decision networks

Summary form only given, as follows. Artificial neural architectures such as the Hopfield network have been applied to optimization problems such as the traveling salesman problem, signal decision problems, and linear programming. However, due to the presence of local minima in the energy function of Hopfield networks, the percentage of incorrect outputs increases as the problem size becomes larger. A design technique based on asymmetrically adjusting the parameters to eliminate such local minima is presented. For the A/D conversion problem, simulation results show perfect A/D conversion with the proposed technique. The technique can be used as a guideline to adjust theoretically derived parameters of signal decision networks in order to obtain the desired output.<>