D. R. Jones
Abstract
We present a new algorithm for finding the global minimum of a multivariate function subject to simple bounds. The algorithm is a modification of the standard Lipschitzian approach that eliminates the need to specify a Lipschitz constant. This is done by carrying out simultaneous searches using all possible constants from zero to infinity. On nine standard test functions, the new algorithm converges in fewer function evaluations than most competing methods.
The motivation for the new algorithm stems from a different way of looking at the Lipschitz constant. In particular, the Lipschitz constant is viewed as a weighting parameter that indicates how much emphasis to place on global versus local search. In standard Lipschitzian methods, this constant is usually large because it must equal or exceed the maximum rate of change of the objective function. As a result, these methods place a high emphasis on global search and exhibit slow convergence. In contrast, the new algorithm carries out simultaneous searches using all possible constants, and therefore operates at both the global and local level. Once the global part of the algorithm finds the basin of convergence of the optimum, the local part of the algorithm quickly and automatically exploits it. This accounts for the fast convergence of the new algorithm on the test functions.
- Stuckman, B., Care, M., andStuckman, P.,System Optimization Using Experimental Evaluation of Design Performance, Engineering Optimization, Vol. 16, pp. 275---289, 1990.Google Scholar
Cross Ref
- Shubert, B.,A Sequential Method Seeking the Global Maximum of a Function, SIAM Journal on Numerical Analysis, Vol. 9, pp. 379---388, 1972.Google ScholarDigital Library
- Galperin, E.,The Cubic Algorithm, Journal of Mathematical Analysis and Applications, Vol. 112, pp. 635---640, 1985.Google ScholarCross Ref
- Pinter, J.,Globally Convergent Methods for n-Dimensional Multiextremal Optimization, Optimization, Vol. 17, pp. 187---202, 1986.Google ScholarCross Ref
- Horst, R., andTuy, H.,On the Convergence of Global Methods in Multiextremal Optimization, Journal of Optimization Theory and Applications, Vol. 54, pp. 253---271, 1987.Google ScholarDigital Library
- Mladineo, R.,An Algorithm for Finding the Global Maximum of a Multimodal, Multivariate Function, Mathematical Programming, Vol. 34, pp. 188---200, 1986. Google ScholarDigital Library
- Preparata, F., andShamos, M.,Computational Geometry: An Introduction, Springer-Verlag, New York, New York, 1985. Google ScholarDigital Library
- Dixon, L., andSzego, G.,The Global Optimization Problem: An Introduction, Toward Global Optimization 2, Edited by L. Dixon and G. Szego, North-Holland, New York, New York, pp. 1---15, 1978.Google Scholar
- Yao, Y.,Dynamic Tunneling Algorithm for Global Optimization, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 19, pp. 1222---1230, 1989.Google ScholarCross Ref
- Stuckman, B., andEason, E.,A Comparison of Bayesian Sampling Global Optimization Techniques, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 22, pp. 1024---1032, 1992.Google ScholarCross Ref
- Belisle, C., Romeijn, H., andSmith, R.,Hide-and-Seek: A Simulated Annealing Algorithm for Global Optimization, Technical Report 90-25, Department of Industrial and Operations Engineering, University of Michigan, 1990.Google Scholar
- Boender, C., et al.,A Stochastic Method for Global Optimization, Mathematical Programming, Vol. 22, pp. 125---140, 1982. Google ScholarDigital Library
- Snyman, J., andFatti, L.,A Multistart Global Minimization Algorithm with Dynamic Search Trajectories, Journal of Optimization Theory and Applications, Vol. 54, pp. 121---141, 1987.Google ScholarDigital Library
- Kostrowicki, J., andPiela, L.,Diffusion Equation Method of Global Minimization: Performance on Standard Test Functions, Journal of Optimization Theory and Applications, Vol. 69, pp. 269---284, 1991. Google ScholarDigital Library
- Perttunen, C.,Global Optimization Using Nonparametric Statistics, University of Louisville, PhD Thesis, 1990. Google ScholarDigital Library
- Perttunen, C., andStuckman, B.,The Rank Transformation Applied to a Multiunivariate Method of Global Optimization, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 20, pp. 1216---1220, 1990.Google ScholarCross Ref
Index Terms
Lipschitzian optimization without the Lipschitz constant
Recommendations
A branch and bound algorithm for the global optimization of Hessian Lipschitz continuous functions
We present a branch and bound algorithm for the global optimization of a twice differentiable nonconvex objective function with a Lipschitz continuous Hessian over a compact, convex set. The algorithm is based on applying cubic regularisation techniques ...
Branching and bounding improvements for global optimization algorithms with Lipschitz continuity properties
We present improvements to branch and bound techniques for globally optimizing functions with Lipschitz continuity properties by developing novel bounding procedures and parallelisation strategies. The bounding procedures involve nonconvex quadratic or ...
Optimal algorithms for global optimization in case of unknown Lipschitz constant
Special issue: Algorithms and complexity for continuous problems Schloss Dagstuhl, Germany, September 2004We consider the global optimization problem for d-variate Lipschitz functions which, in a certain sense, do not increase too slowly in a neighborhood of the global minimizer(s). On these functions, we apply optimization algorithms which use only ...
Comments