Iterative Topographical Global Optimization
In topographical global optimization a sample of points that super-uniformly cover the region of interest, A, is used in combination with the function evaluations f(x) in these points to obtain a topographical graph of/on A from which candidate points are easily extracted for local minimizations. This paper discusses some of the problems in obtaining such a cover and presents some solutions. These solutions are based on an iterative use of the topographical method. Several iterations of the topographical algorithm are run and the information gathered is collected into a single graph. Using multiple iterations speeds up the sampling process and also allows using the topographical method for constrained problems.
KeywordsGlobal optimization uniform random sampling
Unable to display preview. Download preview PDF.
- 2.L. C. W. Dixon and G. P. Szegö (Eds.), Towards Global Optimization 2, North-Holland, Amsterdam, 1978, 363 pp.Google Scholar
- 3.H. Niederreiter, Random Number Generation and Quasi-Monte Carlo Methods, CBMS-NSF regional conference series in applied mathematics 63, Society for Industrial and Applied Mathematics, Philadelphia, Pennsylvania, 1993, 241 pp.Google Scholar
- 4.S. Tezuka and P. L’Ecuyer, Efficient and Portable Combined Tausworthe Random Number Generators, In: ACM Transactions on Modelling and Computer Simulation, Vol. 1, No. 2, April 1991, 99–112.Google Scholar
- 5.A. Torn and S. Viitanen, Topographical Global Optimization, In: C. A. Floudas and P. M. Pardalos ( Eds. ), Recent Advances in Global Optimization (Princeton University Press), 633 p.Google Scholar