TERMINAL REPELLER UNCONSTRAINED SUBENERGY TUNNELING (TRUST) FOR FASTGLOBAL OPTIMIZATION

A new method for unconstrained global function optimization, acronymed TRUST, is introduced. This method formulates optimization as the solu tion of a deterministic dynamical system incorporating terminal repell ers and a novel subenergy tunneling function. Benchmark tests comparin g this method to other global optimization procedures are presented, a nd the TRUST algorithm is shown to be substantially faster. The TRUST formulation leads to a simple stopping criterion. In addition, the str ucture of the equations enables an implementation of the algorithm in analog VLSI hardware, in the vein of artificial neural networks, for f urther substantial speed enhancement.