Parsimony Pressure versus Multi-objective Optimization for Variable Length Representations
We contribute to the theoretical understanding of variable length evolutionary algorithms. Such algorithms are very flexible but can encounter the bloat problem which means solutions grow during the optimization run without providing additional benefit. We explore two common mechanisms for dealing with this problem from a theoretical point of view and point out the differences of a parsimony and a multi-objective approach in a rigorous way. As an example to point out the differences, we consider different measures of sortedness for the classical sorting problem which has already been studied in the computational complexity analysis of evolutionary algorithms with fixed length representations.
KeywordsEvolutionary Algorithm Pareto Front Mutation Operator Correct Position Length Representation
Unable to display preview. Download preview PDF.
- 1.Auger, A., Doerr, B. (eds.): Theory of Randomized Search Heuristics: Foundations and Recent Developments. World Scientific (2011)Google Scholar
- 3.Durrett, G., Neumann, F., O’Reilly, U.-M.: Computational complexity analysis of simple genetic programing on two problems modeling isolated program semantics. In: FOGA, pp. 69–80. ACM (2011)Google Scholar
- 6.Kötzing, T., Sutton, A., Neumann, F., O’Reilly, U.-M.: The Max problem revisited: the importance of mutation in genetic programming. In: GECCO (to appear, 2012)Google Scholar
- 9.Neumann, F.: Computational complexity analysis of multi-objective genetic programming. In: GECCO (to appear, 2012), http://arxiv.org/abs/1203.4881
- 10.Neumann, F., Wegener, I.: Minimum spanning trees made easier via multi-objective optimization. In: GECCO, pp. 763–770. ACM Press (2005)Google Scholar
- 11.Neumann, F., Witt, C.: Bioinspired Computation in Combinatorial Optimization – Algorithms and Their Computational Complexity. Springer (2010)Google Scholar
- 14.Urli, T., Wagner, M., Neumann, F.: Experimental Supplements to the Computational Complexity Analysis of Genetic Programming for Problems Modelling Isolated Program Semantics. In: Coello Coello, C.A., et al. (eds.) PPSN 2012, Part I. LNCS, vol. 7491, pp. 102–112. Springer, Heidelberg (2012)Google Scholar
- 15.Wegener, I.: Methods for the analysis of evolutionary algorithms on pseudo-boolean functions. In: Evolutionary Optimization. International Series in Operations Research and Management Science, vol. 48, pp. 349–369. Springer, US (2003)Google Scholar