Guiding Single-Objective Optimization Using Multi-objective Methods
This paper investigates the possibility of using multi-objective methods to guide the search when solving single-objective optimization problems with genetic algorithms.Using the job shop scheduling problem as an example,experiments demonstrate that by using helper-objectives (additional objectives guiding the search),the average performance of a standard GA can be significantly improved.The helper-objectives guide the search towards solutions containing good building blocks and helps the algorithm avoid local optima.The experiments reveal that the approach only works if the number of helper-objectives used simultaneously is low.However,a high number of helper-objectives can be used in the same run by changing the helper-objectives dynamically.
KeywordsProblem Instance Multiobjective Optimization Traditional Algorithm Travelling Salesperson Problem Good Building Block
Unable to display preview. Download preview PDF.
- 2.S. Bleuler, M. Brack, L. Thiele, and C. Zitzler. Multiobjective Genetic Programming:Reducing Bloat using SPEA2.In Proceedings of CEC’ 2001,pages 536–543, 2001.Google Scholar
- 3.D. Corne, N. Jerram, J. Knowles, and M. Oates. PESA-II:Region-based Selection in Evolutionary Multiobjective Optimization. In L. Spector et al., editors, Proceedings of GECCO-2001:Genetic and Evolutionary Computation Conference, pages 283–290. Morgan Kaufmann, 2001.Google Scholar
- 4.E. D. de Jong, R. A. Watson, and J. B. Pollack.Reducing Bloat and Promoting Diversity using Multi-Objective Methods. In L. Spector et al., editors, Proceedings of GECCO’ 2001, pages 11–18. Morgan Kaufmann, 2001.Google Scholar
- 6.H. Fisher and G. L. Thompson. Probabilistic learning combinations of local job-shop scheduling rules.In J. F. Muth and G. L. Thompson, editors, Industrial Scheduling, pages 225–251. Prentice Hall, 1963.Google Scholar
- 8.M. T. Jensen. Robust and Flexible Scheduling with Evolutionary Computation. PhD thesis, Department of Computer Science, University of Aarhus, 2001.Google Scholar
- 9.M. T. Jensen. Reducing the Run-time Complexity of the NSGA-II. In submission, 2002. Currently available from http://www.daimi.au.dk/~mjensen/.
- 10.J. D. Knowles, R. A. Watson, and D. W. Corne. Reducing Local Optima in Single-Objective Problems by Multi-objectivization.In E. Zitzler et al., editors, Proceedings of the First International Conference on Evolutionary Multi-criterion Optimization (EMO’ 01), pages 269–283. Springer-Verlag, 2001.Google Scholar
- 11.S. Lawrence. Resource constrained project scheduling:an experimental investigation of heuristic scheduling techniques (Supplement). Graduate School of Industrial Administration, Carnegie-Mellon University, 1984.Google Scholar
- 12.S. J. Louis and G. J. E. Rawlins. Pareto Optimality, GA-easiness and Deception. In S. Forrest, editor, Proceedings of ICGA-5, pages 118–123. Morgan Kaufmann, 1993.Google Scholar
- 13.J. Noble and R. A. Watson. Pareto coevolution:Using performance against coevolved opponents in a game as dimensions for Pareto selection. In L. Spector et al., editors, Proceedings of GECCO’ 2001, pages 493–500. Morgan Kaufmann, 2001.Google Scholar
- 14.J. S. Scharnow, K. Tinnefeld, and I. Wegener. Fitness Landscapes Based on Sorting and Shortest Paths Problems. In J. J. Merelo Guerv ós et al., editors, Proceedings of PPSN VII, volume 2439 of LNCS, pages 54–63. Springer-Verlag, 2002.Google Scholar
- 15.R. A. Watson and J. B. Pollack. Symbiotic Combination as an Alternative to Sexual Recombination in Genetic Algorithms. In M. Schoenauer et al., editors, Proceedings of PPSN VI, volume 1917 of LNCS, pages 425–434. Springer-Verlag, 2000.Google Scholar