Predicting Good Propagation Methods for Constraint Satisfaction
Given the breadth of constraint satisfaction problems (CSPs) and the wide variety of CSP solvers, it can be difficult to determine a priori which solving method is best suited to a problem. We explore the use of machine learning to predict which solving method will be most effective for a given problem. Our investigation studies the problem of attribute selection for CSPs, and supervised learning to classify CSP instances drawn from four distinct CSP classes. We limit our study to the choice of two well-known, but simple, CSP solvers. We show that the average performance of the resulting solver is very close to the average performance of a CSP solver based on an oracle.
Unable to display preview. Download preview PDF.
- 1.Régin, J.: A filtering algorithm for constraints of difference in csps. In: Proceedings of the National Conference on Artificial Intelligence, p. 362. John Wiley & Sons Ltd. (1994)Google Scholar
- 2.Puget, J.: A fast algorithm for the bound consistency of alldiff constraints. In: Proceedings of the National Conference on Artificial Intelligence, pp. 359–366. John Wiley & Sons Ltd. (1998)Google Scholar
- 4.Sabin, D., Freuder, E.: Contradicting conventional wisdom in constraint satisfaction. In: Proceedings of the 11th European Conference on Artificial Intelligence, pp. 125–129. Springer, Heidelberg (1994)Google Scholar
- 9.Epstein, S., Wallace, R., Freuder, E., Li, X.: Learning propagation policies. In: Proceedings of the Second International Workshop on Constraint Propagation And Implementation, pp. 1–15 (2005)Google Scholar
- 12.Stergiou, K.: Heuristics for dynamically adapting propagation. In: Proceedings of the 18th European Conference on Artificial Intelligence, pp. 485–489. IOS Press (2008)Google Scholar
- 16.Gomes, C., Shmoys, D.: Completing quasigroups or latin squares: A structured graph coloring problem. In: Proceedings of the Computational Symposium on Graph Coloring and Generalizations, pp. 22–39 (2002)Google Scholar
- 17.Thompson, C.D.S.: Metareasoning about propagators for constraint satisfaction. Master’s thesis, Department of Computer Science, University of Saskatchewan (2011)Google Scholar
- 18.Hall, M.: Correlation-based feature selection for discrete and numeric class machine learning. In: Proceedings of the 17th International Conference on Machine Learning, pp. 359–366 (2000)Google Scholar
- 19.Gent, I., MacIntyre, E., Prosser, P., Walsh, T.: The constrainedness of search. In: Proceedings of the 13th National Conference on Artificial Intelligence, pp. 246–252 (1996)Google Scholar