Landscape State Machines: Tools for Evolutionary Algorithm Performance Analyses and Landscape/Algorithm Mapping
Many evolutionary algorithm applications involve either fitness functions with high time complexity or large dimensionality (hence very many fitness evaluations will typically be needed) or both. In such circumstances, there is a dire need to tune various features of the algorithm well so that performance and time savings are optimized. However, these are precisely the circumstances in which prior tuning is very costly in time and resources. There is hence a need for methods which enable fast prior tuning in such cases. We describe a candidate technique for this purpose, in which we model a landscape as a finite state machine, inferred from preliminary sampling runs. In prior algorithm-tuning trials, we can replace the ‘real’ landscape with the model, enabling extremely fast tuning, saving far more time than was required to infer the model. Preliminary results indicate much promise, though much work needs to be done to establish various aspects of the conditions under which it can be most beneficially used. A main limitation of the method as described here is a restriction to mutationonly algorithms, but there are various ways to address this and other limitations.
KeywordsMutation Operator Finite State Machine Algorithm Performance Fitness Evaluation Fitness Landscape
Unable to display preview. Download preview PDF.
- 1.Altenberg, L. Fitness distance correlation analysis: an instructive counterexample. In Th. Bäck, editor, Proceedings of the 7th International Conference on Genetic Algorithms, pages 57–64. Morgan Kaufmann Publishers, 1997.Google Scholar
- 4.Barnett, L. Ruggedness and neutrality: the NKp family of fitness landscapes. In C. Adami, R. K. Belew, H. Kitano, and C. E. Taylor, editors, Alive VI: Sixth International Conference on Articial Life, pages 18–27, Cambridge MA, 1998. MIT Press.Google Scholar
- 5.Davidor, Y. (1991): “Epistasis Variance: A Viewpoint on GA-Hardness”. In: Foundations of genetic algorithms, ed. G.J.E. Rawlins, Morgan Kaufmann Publishers, pp. 23–35.Google Scholar
- 6.Grefenstette, J.J. (1992). Deception considered harmful. Foundations of Genetic Algorithms, 2. Whitley, L. D., (ed.), Morgan Kaufmann, 75–91.Google Scholar
- 7.Jones, T. and S. Forrest. Fitness Distance Correlation as a Measure of Problem Difficulty for Genetic Algorithms. In L. J. Eshelman, editor, Proceedings of the 6th Int. Conference on Genetic Algorithms, pages 184–192, Kaufman, 1995.Google Scholar
- 9.Kallel, L., Naudts, B. & Reeves, C. (1998). Properties of fitness functions and search landscapes. In Theoretical aspects of evolutionary computing (ed. L. Kallel, B. Naudts and A. Rogers), pp. 175–206. Springer, Berlin.Google Scholar
- 12.Stadler, P.F. Towards a Theory of Landscapes,“ in Complex Systems and Binary Networks,” (R. Lopez-Pena et al, eds.), Berlin, New York, pp. 77–163, Springer Verlag, 1995.Google Scholar
- 14.Wright, S. (1932). The roles of mutation, inbreeding, crossbreeding and selection in evolution. In Proc. Sixth Int. Conf. Genetics, vol. 1 (ed. D. F. Jones), pp. 356–366.Google Scholar