Globalization and Parallelization of Nelder-Mead and Powell Optimization Methods
- 670 Downloads
Optimization problems in engineering involve very often nonlinear functions with multiple minima or discontinuities, or the simulation of a system in order to determine its parameters. Global search methods can compute a set of points and provide alternative design answers to a problem, but are computationally expensive. A solution for the CPU dependency is parallelization, which leads to the need to control the sampling of the search space. This paper presents a parallel implementation of two free-derivative optimization methods (Nelder-Mead and Powell), combined with two restart strategies to globalize the search. The first is based on a probability density function, while the second uses a fast algorithm to uniformly sample the space. The implementation is suited to a faculty network, avoiding special hardware requirements, complex installation or coding details.
Unable to display preview. Download preview PDF.
- M. Mitchell,An Introduction to Genetic Algorithms (Complex Adaptive Systems): The MIT Press, 1998.Google Scholar
- P. Moscato, “On Evolution, Search, Optimization, Genetic Algorithms and Martial Arts: Towards Memetic Algorithms.,” Caltech Concurrent Computation Program C3P Report 826, 1989.Google Scholar
- L. Bottou and N. Murata, “Stochastic Approximations and Efficient Learning,” inThe Handbook of Brain Theory and Neural Networks, M. A. Arbib, Ed. Cambridge, USA.: Massachusetts Institute of Technology, 2002.Google Scholar
- E. Cantu-Paz, “A survey of parallel genetic algorithms.,”Calculateurs Parallèles, Reseaux et Systèmes Répartis., vol. 10, pp. 141-171, 1998.Google Scholar
- A. Bürmen, J. Puhan, T. Tuma, I. Fajfar, and A. Nussdorfer, “Parallel simplex algorithm for circuit optimisation,” inERK 2001, Portorož, Slovenija, 2001.Google Scholar
- A. Lewis, “Parallel Optimisation Algorithms for Continuous Non-Linear Numerical Simulations,” inFaculty of Engineering and Information Technology. vol. Doctor Brisbane, Australia: Griffith University, 2004.Google Scholar
- R. Lewis, V. Torczon, and M. Trosset, “Direct search methods: then and now,”Journal of Computational and Applied Mathematics, vol. 124, 2000.Google Scholar
- R. P. Brent, “Algorithms for Minimization without Derivatives,v inAlgorithms for Minimization without Derivatives Englewood Cliffs, NJ, EUA: Prentice-Hall, 1973.Google Scholar
- W. T. Vetterling and B. P. Flannery,Numerical Recipes in C++: The Art of Scientific Computing: Cambridge University Press, 2002.Google Scholar
- D. Marshall, “Nearest Neighbour Searching in High Dimensional Metric Spaces,” inDepartment of Computer Sciences. vol. Master Thesis in Information Technology: Australian National University, 2006.Google Scholar
- S. A. Nene and S. K. Nayar, “A simple algorithm for nearest neighbor search in high dimensions,”IEEE Transactions on Pattern Analysis and Machine Intelligence,pp. 989-1003, 1997.Google Scholar
- J. E. Souza de Cursi and A. Koscianski, “Physically Constrained Neural Network Models for Simulation,” inAdvances and Innovations in Systems, Computing, Sciences and Software Engineering, K. Elleithy, Ed. Dordrecht, The Netherlands: Springer, 2007.Google Scholar