Multiple Parallel Local Searches in Global Optimization

The unconstrained global programming problem is addressed using an efficient multi-start algorithm, in which parallel local searches contribute towards a Bayesian global stopping criterion. The stopping criterion, denoted the unified Bayesian global stopping criterion, is based on the mild assumption that the probability of convergence to the global optimum x* is comparable to the probability of convergence to any local minimum xj. The combination of the simple multi-start local search strategy and the unified Bayesian global stopping criterion outperforms a number of leading global optimization algorithms, for both serial and parallel implementations. Results for parallel clusters of up to 128 machines are presented.

[1]  Alexander H. G. Rinnooy Kan,et al.  A stochastic method for global optimization , 1982, Math. Program..

[2]  Aimo A. Törn,et al.  Global Optimization , 1999, Science.

[3]  Parisi,et al.  SIGMA - A Stochastic-Integration Global Minimization Algorithm. , 1985 .

[4]  F. Aluffi-Pentini,et al.  Global optimization and stochastic differential equations , 1985 .

[5]  Jorge Nocedal,et al.  Global Convergence Properties of Conjugate Gradient Methods for Optimization , 1992, SIAM J. Optim..

[6]  Francesco Zirilli,et al.  Algorithm 667: Sigma—a stochastic-integration global minimization algorithm , 1988, TOMS.

[7]  C. G. E. Boender,et al.  A Bayesian Analysis of the Number of Cells of a Multinomial Distribution , 1983 .

[8]  Jorge Nocedal,et al.  A Limited Memory Algorithm for Bound Constrained Optimization , 1995, SIAM J. Sci. Comput..

[9]  G. T. Timmer,et al.  Stochastic global optimization methods part I: Clustering methods , 1987, Math. Program..

[10]  Fabio Schoen,et al.  Stochastic techniques for global optimization: A survey of recent advances , 1991, J. Glob. Optim..

[11]  J. Snyman,et al.  A multi-start global minimization algorithm with dynamic search trajectories , 1987 .

[12]  A. Griewank.  Generalized descent for global optimization , 1981 .

[13]  Jorge Nocedal,et al.  Algorithm 778: L-BFGS-B: Fortran subroutines for large-scale bound-constrained optimization , 1997, TOMS.

[14]  J. Mockus,et al.  The Bayesian approach to global optimization , 1989 .