Felix Dobslaw
ABSTRACT
That there is no best initial parameter setting for a metaheuristic on all optimization problems is a proven fact (no free lunch theorem). This paper studies the applicability of so called robust parameter settings for combinatorial optimization problems. Design of Experiments supported parameter screening had been carried out, analyzing a discrete Particle Swarm Optimization algorithm on three demographically very dissimilar instances of the Traveling Salesmen Problem. First experimental results indicate that parameter settings produce varying performance quality for the three instances. The robust parameter setting is outperformed in two out of three cases. The results are even significantly worse when considering quality/time trade-off. A methodology for problem generalization is referred to as a possible solution.
- T. Bartz-Beielstein, C. Lasarczyk, and M. Preuss. Sequential parameter optimization. volume 1, pages 773--780, 2005.Google Scholar
- H.-. Beyer and B. Sendhoff. Robust optimization - a comprehensive survey. Computer Methods in Applied Mechanics and Engineering, 196(33-34):3190--3218, 2007.Google ScholarCross Ref
- M. Cunkas and M. Y. Özsaglam. A comparative study on particle swarm optimization and genetic algorithms for traveling salesman problems. Cybernetics and Systems, 40(6):490--507, 2009. Google ScholarDigital Library
- F. Dobslaw. A parameter tuning framework for metaheuristics based on design of experiments and artificial neural networks. In Proceedings of ICCMNC 2010, (Rome, Italy, Apr 28 - April 30, 2010), April 2010.Google Scholar
- E. Eiben, R. Hinterding, and Z. Michalewicz. Parameter control in evolutionary algorithms. IEEE Transactions on Evolutionary Computation, 3(2):124--141, 1999. Google ScholarDigital Library
- L. Eriksson, E. Johansson, C. Kettaneh-Wold, and S. W. Wold. Design of Experiments, Principles and Applications. MKS Umetrics AB, 2008.Google Scholar
- J. J. Grefenstette. Optimization of control parameters for genetic algorithms. IEEE Transactions on Systems, Man and Cybernetics, 16(1):122--128, 1986. Google ScholarDigital Library
- F. Hutter, H. H. Hoos, K. Leyton-Brown, and T. Stützle. Paramils: An automatic algorithm configuration framework. Journal of Artificial Intelligence Research, 36:267--306, 2009. Google ScholarCross Ref
- D. Jones, M. Schonlau, and W.J. Welch. Efficient global optimization of expensive black-box functions. Journal of Global Optimization, 13(4):455--492, 1998. Google ScholarDigital Library
- W. A. D. Landgraaf, A. E. Eiben, and V. Nannen. Parameter calibration using meta-algorithms. pages 71--78, 2008.Google Scholar
- J. Marczyk. Stochastic multidisciplinary improvement: beyond optimization. American Institute of Aeronautics and Astronautics, AIAA-200-4929:1--7, 2000.Google Scholar
- Sacks, J. Schiller, S. B., and W. J. Welch. Designs for computer experiments. Technometrics, 31(1):41--47, 1989.Google ScholarCross Ref
- K.-P. Wang, L. Huang, C.-G. Zhou, and W. Pang. Particle swarm optimization for traveling salesman problem. volume 3, pages 1583--1585, 2003.Google Scholar
- D. H. Wolpert and W. G. Macready. No free lunch theorems for optimization. IEEE Transactions on Evolutionary Computation, 1(1):67--82, 1997. Google ScholarDigital Library
Index Terms
- An experimental study on robust parameter settings
Recommendations
Feature-based tuning of simulated annealing applied to the curriculum-based course timetabling problem
We consider the university course timetabling problem, which is one of the most studied problems in educational timetabling. In particular, we focus our attention on the formulation known as the curriculum-based course timetabling problem (CB-CTT), ...
Static and Dynamic Parameter Settings of Accelerated Particle Swarm Optimisation for Solving Course Scheduling Problem
Cooperative Design, Visualization, and EngineeringAbstractThe university course timetabling problem (UCTP) is one of the most challenging scheduling problems and also classified to be a Non-deterministic Polynomial (NP)-hard problem. An Accelerated Particle Swarm Optimisation based Timetabling (APSOT) ...
Parameter tuning for meta-heuristics
AbstractThese days meta-heuristic algorithms are gaining lot of popularity. The performance of the meta-heuristics depends upon the suitable selection of user dependent parameters. Finding the most suitable values for the parameters (fine ...
Graphical abstractDisplay Omitted
Highlights- A generic parameter tuning methodology for meta-heuristic algorithms is proposed.
Comments