Generic Postprocessing via Subset Selection for Hypervolume and Epsilon-Indicator
作者： Karl Bringmann , Tobias Friedrich , Patrick Klitzke
DOI: 10.1007/978-3-319-10762-2_51
关键词: Selection (genetic algorithm) 、 Evolutionary algorithm 、 Multi-objective optimization 、 Evolutionary computation 、 Mathematical optimization 、 Population 、 Process (computing) 、 Computer science 、 Fitness function
摘要: Most biobjective evolutionary algorithms maintain a population of fixed size μ and return the final at termination. During optimization process many solutions are considered, but most discarded. We present two generic postprocessing which utilize archive all non-dominated evaluated during search. choose best from such that hypervolume or e-indicator is maximized. This costs no additional fitness function evaluations has negligible runtime compared to EMOAs.
springer.com
mpg.de
core.ac.uk
sci-hub.st 参考文章(28)
Aníbal Ponte, Luís Paquete, José R. Figueira, On beam search for multicriteria combinatorial optimization problems integration of ai and or techniques in constraint programming. pp. 307- 321 ,(2012) , 10.1007/978-3-642-29828-8_20
Michael Emmerich, Nicola Beume, Boris Naujoks, None, An EMO algorithm using the hypervolume measure as selection criterion international conference on evolutionary multi criterion optimization. pp. 62- 76 ,(2005) , 10.1007/978-3-540-31880-4_5
O. Giel, Expected runtimes of a simple multi-objective evolutionary algorithm congress on evolutionary computation. ,vol. 3, pp. 1918- 1925 ,(2003) , 10.1109/CEC.2003.1299908
Eckart Zitzler, Simon Künzli, Indicator-Based Selection in Multiobjective Search parallel problem solving from nature. pp. 832- 842 ,(2004) , 10.1007/978-3-540-30217-9_84
Karl Bringmann, Tobias Friedrich, Markus Wagner, Frank Neumann, Approximation-guided evolutionary multi-objective optimization international joint conference on artificial intelligence. pp. 1198- 1203 ,(2011) , 10.5591/978-1-57735-516-8/IJCAI11-204
Simon Huband, Luigi Barone, Lyndon While, Phil Hingston, A scalable multi-objective test problem toolkit international conference on evolutionary multi criterion optimization. pp. 280- 295 ,(2005) , 10.1007/978-3-540-31880-4_20
Genetic and Evolutionary Computation - GECCO 2004 Springer Verlag. ,(2004) , 10.1007/B98643
Daniel Vaz, Luís Paquete, Aníbal Ponte, A note on the ε-indicator subset selection Theoretical Computer Science. ,vol. 499, pp. 113- 116 ,(2013) , 10.1016/J.TCS.2013.05.013
Karl Bringmann, Tobias Friedrich, Convergence of hypervolume-based archiving algorithms I Proceedings of the 13th annual conference on Genetic and evolutionary computation - GECCO '11. pp. 457- 464 ,(2011) , 10.1145/2001576.2001678
Anne Auger, Johannes Bader, Dimo Brockhoff, Eckart Zitzler, Investigating and exploiting the bias of the weighted hypervolume to articulate user preferences Proceedings of the 11th Annual conference on Genetic and evolutionary computation - GECCO '09. pp. 563- 570 ,(2009) , 10.1145/1569901.1569980