Simple Rules for Low-Knowledge Algorithm Selection
This paper addresses the question of selecting an algorithm from a predefined set that will have the best performance on a scheduling problem instance. Our goal is to reduce the expertise needed to apply constraint technology. Therefore, we investigate simple rules that make predictions based on limited problem instance knowledge. Our results indicate that it is possible to achieve superior performance over choosing the algorithm that performs best on average on the problem set. The results hold over a variety of different run lengths and on different types of scheduling problems and algorithms. We argue that low-knowledge approaches are important in reducing expertise required to exploit optimization technology.
KeywordsSchedule Problem Problem Instance Algorithm Selection Simple Rule Prediction Rule
Unable to display preview. Download preview PDF.
- 4.Horvitz, E., Ruan, Y., Gomes, C., Kautz, H., Selman, B., Chickering, M.: A bayesian approach to tacking hard computational problems. In: Proceedings of the Seventeenth Conference on uncertainty and Artificial Intelligence (UAI 2001), pp. 235–244 (2001)Google Scholar
- 5.Kautz, H., Horvitz, E., Ruan, Y., Gomes, C., Selman, B.: Dynamic restart policies. In: Proceedings of the Eighteenth National Conference on Artifiical Intelligence (AAAI 2002), pp. 674–681 (2002)Google Scholar
- 7.Watson, J.P.: Empirical Modeling and Analysis of Local Search Algorithms for the Job-Shop Scheduling Problem. PhD thesis, Dept. of Computer Science, Colorado State University (2003)Google Scholar
- 8.Watson, J.P., Barbulescu, L., Whitley, L., Howe, A.: Constrasting structured and random permutation flow-shop scheduling problems: search-space topology and algorithm performance. INFORMS Journal on Computing 14 (2002)Google Scholar
- 11.Nuijten, W.P.M.: Time and resource constrained scheduling: a constraint satisfaction approach. PhD thesis, Department of Mathematics and Computing Science, Eindhoven University of Technology (1994)Google Scholar
- 14.Scheduler: ILOG Scheduler 5.2 User’s Manual and Reference Manual. ILOG, S.A. (2001) Google Scholar
- 15.Beck, J.C., Perron, L.: Discrepancy-bounded depth first search. In: Proceedings of the Second International Workshop on Integration of AI and OR Technologies for Combinatorial Optimization Problems, CPAIOR 2000 (2000)Google Scholar
- 17.Beck, J.C., Refalo, P.: Combining local search and linear programming to solve earliness/tardiness scheduling problems. In: Proceedings of the Fourth International Workshop on Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems, CPAIOR 2002 (2002)Google Scholar
- 18.Vazquez, M., Whitley, L.D.: A comparision of genetic algorithms for the dynamic job shop scheduling problem. In: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2000), pp. 1011–1018. Morgan Kaufmann, San Francisco (2000)Google Scholar