Learning algorithm for chaotic dynamical systems that solve optimization problems

A learning algorithm is introduced for chaotic dynamical systems that solve nonlinear optimization problems. The algorithm controls the asymptotic measure of a chaotic dynamical system that searches for the optimum solution and improves the efficiency of chaotic search dynamics. Using several instances of 1- and 2-dimensional nonlinear optimization problems, the performance of the learning algorithm is demonstrated. We also show that the learning algorithm works as chaotic simulated annealing, which brings about gradual convergence of the chaotic search dynamics to a possible optimum solution. (C) 1998 Scripta Technica, Electron Comm Jpn Pt 3, 82(3): 10-21, 1999.