Parameter Prediction Based on Features of Evolved Instances for Ant Colony Optimization and the Traveling Salesperson Problem

Ant colony optimization performs very well on many hard optimization problems, even though no good worst case guarantee can be given. Understanding the reasons for the performance and the influence of its different parameter settings has become an interesting problem. In this paper, we build a parameter prediction model for the Traveling Salesperson problem based on features of evolved instances. The two considered parameters are the importance of the pheromone values and of the heuristic information. Based on the features of the evolved instances, we successfully predict the best parameter setting for a wide range of instances taken from TSPLIB.

[1]  Daniela Favaretto,et al.  On MAX - MIN Ant System's Parameters , 2006, ANTS Workshop.

[2]  Mario A. Muñoz,et al.  A Meta-learning Prediction Model of Algorithm Performance for Continuous Optimization Problems , 2012, PPSN.

[3]  Gerhard Reinelt,et al.  TSPLIB - A Traveling Salesman Problem Library , 1991, INFORMS J. Comput..

[4]  Holger H. Hoos,et al.  Automated Algorithm Configuration and Parameter Tuning , 2012, Autonomous Search.

[5]  Markus Wagner,et al.  Ant colony optimisation and the traveling salesperson problem: hardness, features and parameter settings , 2013, GECCO '13 Companion.

[6]  Vincenzo Cutello,et al.  Parallel Problem Solving from Nature - PPSN XII , 2012, Lecture Notes in Computer Science.

[7]  Daniel Kudenko,et al.  Determining Whether a Problem Characteristic Affects Heuristic Performance , 2008, Recent Advances in Evolutionary Computation for Combinatorial Optimization.

[8]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[9]  Jano I. van Hemert,et al.  Understanding TSP Difficulty by Learning from Evolved Instances , 2010, LION.

[10]  Thomas Stützle,et al.  Off-line vs. on-line tuning: a study on MAX–MIN ant system for the TSP , 2010, ICSI 2010.

[11]  Eric Monfroy,et al.  Autonomous Search , 2012, Springer Berlin Heidelberg.

[12]  Thomas Stützle,et al.  A short convergence proof for a class of ant colony optimization algorithms , 2002, IEEE Trans. Evol. Comput..

[13]  F. Wilcoxon.  Individual Comparisons by Ranking Methods , 1945 .

[14]  Marco Dorigo,et al.  Ant colony optimization , 2006, IEEE Computational Intelligence Magazine.

[15]  Thomas Stützle,et al.  An Analysis of the Hardness of TSP Instances for Two High-performance Algorithms , 2005 .

[16]  Bernd Bischl,et al.  A feature-based comparison of local search and the christofides algorithm for the travelling salesperson problem , 2013, FOGA XII '13.

[17]  Bernd Bischl,et al.  A novel feature-based approach to characterize algorithm performance for the traveling salesperson problem , 2012, Annals of Mathematics and Artificial Intelligence.

[18]  David W. Aha,et al.  Instance-Based Learning Algorithms , 1991, Machine Learning.

[19]  Thomas Stützle,et al.  Ant Colony Optimization and Swarm Intelligence, 6th International Conference, ANTS 2008, Brussels, Belgium, September 22-24, 2008. Proceedings , 2008, ANTS Conference.

[20]  Thomas Stützle,et al.  Off-line vs. On-line Tuning: A Study on MAX-MIN\mathcal{MAX--MIN} Ant System for the TSP , 2010, ANTS Conference.

[21]  Thomas Stützle,et al.  Parameter Adaptation in Ant Colony Optimization , 2012, Autonomous Search.

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

[23]  William J. Cook,et al.  Solution of a Min-Max Vehicle Routing Problem , 2002, INFORMS Journal on Computing.

[24]  Frank Neumann,et al.  Theoretical analysis of two ACO approaches for the traveling salesman problem , 2011, Swarm Intelligence.