A Racing Algorithm for Configuring Metaheuristics

This paper describes a racing procedure for finding, in a limited amount of time, a configuration of a metaheuristic that performs as good as possible on a given instance class of a combinatorial optimization problem. Taking inspiration from methods proposed in the machine learning literature for model selection through cross-validation, we propose a procedure that empirically evaluates a set of candidate configurations by discarding bad ones as soon as statistically sufficient evidence is gathered against them. We empirically evaluate our procedure using as an example the configuration of an ant colony optimization algorithm applied to the traveling salesman problem. The experimental results show that our procedure is able to quickly reduce the number of candidates, and allows to focus on the most promising ones.

[1]  Margaret J. Robertson,et al.  Design and Analysis of Experiments , 2006, Handbook of statistics.

[2]  H. J. Larson.  Introduction to Probability Theory and Statistical Inference , 1970 .

[3]  W. J. Conover,et al.  Practical Nonparametric Statistics , 1972 .

[4]  P. Billingsley,et al.  Probability and Measure , 1980 .

[5]  Patrick Billingsley,et al.  Probability and Measure. , 1986 .

[6]  Jon Jouis Bentley,et al.  Fast Algorithms for Geometric Traveling Salesman Problems , 1992, INFORMS J. Comput..

[7]  Andrew W. Moore,et al.  Hoeffding Races: Accelerating Model Selection Search for Classification and Function Approximation , 1993, NIPS.

[8]  Gerald DeJong,et al.  Learning Search Control Knowledge for Deep Space Network Scheduling , 1993, ICML.

[9]  G. Reinelt.  The traveling salesman: computational solutions for TSP applications , 1994 .

[10]  Andrew W. Moore,et al.  Efficient Algorithms for Minimizing Cross Validation Error , 1994, ICML.

[11]  Jonathan Gratch,et al.  On the Efficient Allocation of Resources for Hypothesis Evaluation: A Statistical Approach , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Holger H. Hoos,et al.  Improving the Ant System: A Detailed Report on the MAX-MIN Ant System , 1996 .

[13]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[14]  T. Stützle,et al.  MAX-MIN Ant System and local search for the traveling salesman problem , 1997, Proceedings of 1997 IEEE International Conference on Evolutionary Computation (ICEC '97).

[15]  E. D. Taillard,et al.  Ant Systems , 1999 .

[16]  Luca Maria Gambardella,et al.  Ant Algorithms for Discrete Optimization , 1999, Artificial Life.

[17]  P. Good.  Resampling Methods , 1999, Birkhäuser Boston.

[18]  Beckman,et al.  Learning as Searc , 1999 .

[19]  Marco Dorigo,et al.  The ant colony optimization meta-heuristic , 1999 .

[20]  Future Generation Computer Systems , 2000 .

[21]  Thomas Stützle,et al.  MAX-MIN Ant System , 2000, Future Gener. Comput. Syst..

[22]  Thomas Stützle,et al.  The Ant Colony Optimization Metaheuristic: Algorithms, Applications, and Advances , 2003 .