SIMULATED ANNEALING USING COARSE-GRAINED CLASSICAL DYNAMICS - SMOLUCHOWSKI DYNAMICS IN THE GAUSSIAN DENSITY APPROXIMATION

A dynamical annealing algorithm for global optimization based on appro ximate solution of the Smoluchowski equation is presented. The equatio ns of motion in the Gaussian density approximation are interpreted as a steepest descent quench on a time dependent effective potential ener gy surface. A relation between the convexity condition for the effecti ve potential surface and the size of thermal fluctuations provides a d efinition of the critical temperature above which the distribution is delocalized and the effective potential is smooth and convex during an annealing run. This critical temperature may be significantly less th an the temperature characteristic of escape from a local energy minimu m. (C) 1995 American Institute of Physics.