Abstract
The performance of a multi-objective evolutionary algorithm (MOEA) is evaluated with regard to the quality of the populations under two aspects: the distance of the non-dominated set of a population to the true Pareto front (\(PF_{true}\)) and the spread among these solutions. Diverse convergence indicators have been proposed in the past with different requirements: either \(PF_{true}\) or a reference set of Pareto-optimal solutions is required. Furthermore, most of the convergence indicators are restricted to a non-dominated solution set, and therefore, the quality of the entire population is only represented by the non-dominated solutions. This work presents a statistically reasonable convergence indicator that is able to reflect the quality of the entire population. The average cuboid volume (ACV) assigns desirable aspects regarding the classification of entire populations. These preferable features are demonstrated and discussed. Furthermore, ACV is used as selection criterion to determine the solutions of the succeeding generation in a proposed customized NSGA-II for biochemical optimization. Two selection strategies based on the ACV indicator are proposed and empirically compared to a Pareto rank-based selection strategy. These selection strategies depend on two parameters and the adaption of the selection pressure by a variation of these parameters is empirically investigated on a three-dimensional biochemical minimization problem.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Okabe, T., Jin, Y., Sendhoff, B.: A critical survey of performance indices for multi-objective optimisation. In: Proceedings of the IEEE Congress on Evolutionary Computation, vol. 2, pp. 878885 (2003)
Grosan, C., Oltean, M., Dumitrescu, D.: Performance metrics for multiobjective optimization evolutionary algorithms. In: Proceedings of the Applied and Industrial Mathematics(2003)
Rosenthal, S., Borschbach, M.: Impact of population size and selection within a customized NSGA-II for biochemical optimization assessed on the basis of the average cuboid volume indicator. In: 6th International Conference on Bioinformatics, Computational Systems and Biotechnologies (BIOTECHNO 2014), IARIA, pp. 1–7 (2014)
Emmerich, M., Beume, N., Naujoks, B.: An EMO Algorithm Using the Hypervolume Measure as Selection Criterion. EMO 2005, LNCS 3410, 62–76 (2005)
Zitzler, E., Künzli, S.: Indicator-based selection in multiobjective search. Parallel Prob. Solving Nat. (PPSN VIII) 3242, 832–842 (2004)
Rosenthal, S., El-Sourani, N., Borschbach, M.: Introduction of a Mutation Specific Fast Non-dominated Sorting GA Evolved for Biochemical Optimization. SEAL 2012, LNCS 7673, 158–167 (2012)
Rosenthal, S., El-Sourani, N., Borschbach, M.: Impact of Different Recombination Methods in a Mutation-Specific MOEA for a Biochemical Application. In: Vanneschi, L., Bush, W.S., Giacobini, M. (eds.) EvoBIO 2013, LNCS 7833, 188–199 (2013)
Rosenthal, S., Borschbach, M.: A benchmark on the interaction of basic variation operators in multi-objective peptide design evaluated by a three dimensional diversity metric and a minimized hypervolume. In: Emmerich, M. et. al. (eds.), EVOLVE - A Bridge between Probability, Set Oriented Numerics and Evolutionary Computation IV, pp. 139–153 (2013)
Hansen, M.P., Jaszkiewicz, A.: Evaluating the Quality of Approximations to the Non-dominated Set. Technical Report IMM-REP-1998-7, Technical University of Denmark (1998)
Knowles, J., Corne, D.: On metrics for comparing nondominated sets. In: Congress on Evolutionary Computation (CEC 2002), pp. 711–716. IEEE Press, New Jersey (2002)
Zitzler, E., Thiele L.: Multiobjective optimization using evolutionary algorithms—a comparative case study. In: Eiben, A.E., Bäck, T., Schoenauer, M., Schwefel, H.P. (eds.), Fifth International Conference on Parallel Problem Solving form Nature (PPSN-V), pp. 292–301. Berlin, Germany (1998)
Zitzler, E.: Evolutionary algorithms for multiobjective optimization: methods and applications. Ph.D. dissertation, Swiss Federal Institute of Technology (ETH) Zurich (1999)
Laumanns, L., Zitzler, E., Thiele, L.: A unified model for multi-objective evolutionary algorithms with elitism. In: CEC 2000, vol. 1, pp. 46–52 (2000)
Zitzler, E., Thiele, L., Laumanns, M., Fonseca, C.M., da Fonseca, V.G.: Performance assessment of multiobjective optimizers: An analysis and review. IEEE Trans. Evolut. Comput. 7(2), 117–132 (2003)
Beume, N., Rudolph, G.: Faster S-metric calculation by considering dominated hypervolume as klees measure problem. Evol. Comput. 17(4), 477–492 (2009)
Beume N., Rudolph, G.: Faster S-metric calculation by considering dominated hypervolume as Klee’s measure problem. In: Proceedings of the Second IASTED Conference on Computational Intelligence, pp. 231–236 (2006)
Bradstreet, L., While, L., Barone, L.: A fast incremental hypervolume algorithm. IEEE Trans. Evolut. Comput. 12(6), 714–723 (2008)
While, L., Bradstreet, L., Barone, L.A.: Fast way of calculating exact hypervolume. IEEE Trans. Evolut. Comput. 10, 29–38 (2006)
Van Veldhuizen, D.A.: Multiobjective evolutionary algorithms: classification, analyses and new innovations. Ph.D. dissertation, Air Force Institute of Technology, Dayton, Ohio (1999)
Veldhuizen, D.A., Lamont, G.B.: Multiobjective evolutionary algorithm test. In: Carroll, J., Haddad, H., Oppenheim, D., Bryant, B., Lamont, G.B. (eds.) Proceedings of the 1999 ACM Symposium on Applied Computing, pp. 351–357. San Antonio, Texas (1999)
Deb, K., Jain, S.: Running Performance Metrics for Evolutionary Multiobjective Optimization. Kan GAL Report No. 2002004, Kanpur Genetic Algorithms Laboratory, Indian Institute of Technology Kanpur (2002)
Schütze, O., Esquivel, X., Lara, A., Coello Coello, C.A.: Using the averaged Hausdorff distance as a performance measure in evolutionary multiobjective optimization. IEEE Trans. Evolut. Comput. 16(4), 504–522 (2012)
Coello Coello, C.A., Cruz Cortis, N.: Solving multiobjective optimization problems using an aritifical immune system. Genetic. Program. Evol. Mach. 6(2), 163–190 (2005)
Trautmann, H., Wagner, T., Brockhoff, D.: Focused multiobjective search using R2-indicator-based selection. In: Learning and Intelligent Optimization, pp. 70–74. Springer, Berlin Heidelberg (2013)
Wagner, T., Trautmann, H., Brockhoff, D.: Reference articulation by means of the R2 indicator. In: Evolutionary Multi-criterion Optimization (EMO 2013), vol. 7811, 81–95 (2013)
Brockhoff, D., Wagner, T., Trautmann, H.: On the properties of the R2 indicator. In: Genetic and Evolutionary Computation Conference (GECCO 2012), pp. 465–472 (2012)
von der Lippe, M.: Deskriptive Statistik. Oldenburg Verlag (2006)
Zitzler, E., Brockhoff, D., Thiele, L.: The hypervolume indicator revisited: on the design of pareto compliant indicators via weighted integration. EMO 2007, 862–876 (2007)
Deb, K., Pratap, A., Agarwal, S.: A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans. Evolut. Comput. 6(2), 182–197 (2002)
Rosenthal, S., Borschbach, M.: Impact of population size, selection and multi-parent recombination within a customized NSGA-II for biochemical optimization. Int. J. Adv. Life Sci. IARIA 6(3&4), 310–324 (2014)
Bäck, T., Schütz, M.: Intelligent mutation rate control in canonical genetic algorithm. In: Proceedings of the International Symposium on Methodology for Intelligent systems, pp. 158–167 (1996)
BioJava: CookBook, release 3.0 http://www.biojava.org/wiki/BioJava. Cite January 2014
Cleland, J.L., Langer, R., Washington, D.C.: Formulation and Delivery of Proteins and Peptides: Design and Development Strategies, pp. 1–19. American Chemical Society (1994)
Hopp, T.P., Woods, K.R.: A computer program for predicting protein antigenic determinants. Mol. Immunol. 20(4), 483–489 (1983)
Hansch, C., Björkroth, J., Leo, A.: Hydrophobicity and central nervous system agents: on the principle of minimal hydrohpobicity in drug design. J. Pharmacol. Sci. 76(9), 663–687 (1987)
Rosenthal, S., Freisleben, B., Borschbach, M.: Aggregate selection in multi- objective biochemical optimization via the average cuboid volume indicator. In: Emmerich et al. (eds.) EVOLVE—A Bridge between Probability, Set Oriented Numerics and Evolutionary Computation VI (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this chapter
Cite this chapter
Rosenthal, S., Borschbach, M. (2017). Average Cuboid Volume as a Convergence Indicator and Selection Criterion for Multi-objective Biochemical Optimization. In: Emmerich, M., Deutz, A., Schütze, O., Legrand, P., Tantar, E., Tantar, AA. (eds) EVOLVE – A Bridge between Probability, Set Oriented Numerics and Evolutionary Computation VII. Studies in Computational Intelligence, vol 662. Springer, Cham. https://doi.org/10.1007/978-3-319-49325-1_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-49325-1_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-49324-4
Online ISBN: 978-3-319-49325-1
eBook Packages: EngineeringEngineering (R0)