Integrating User Preferences into Evolutionary Multi-Objective Optimization

Many real-world optimization problems involve multiple, typically conflicting objectives. Often, it is very difficult to weigh the different criteria exactly before alternatives are known. Evolutionary multi-objective optimization usually solves this predicament by searching for the whole Pareto-optimal front of solutions. However, often the user has at least a vague idea about what kind of solutions might be preferred. In this chapter, we argue that such knowledge should be used to focus the search on the most interesting (from a user’s perspective) areas of the Paretooptimal front. To this end, we present and compare two methods which allow to integrate vague user preferences into evolutionary multi-objective algorithms. As we show, such methods may speed up the search and yield a more fine-grained selection of alternatives in the most relevant parts of the Pareto-optimal front.

[1]  Gary B. Lamont,et al.  Multiobjective Evolutionary Algorithms: Analyzing the State-of-the-Art , 2000, Evolutionary Computation.

[2]  Kalyanmoy Deb,et al.  Multi-objective evolutionary algorithms: introducing bias among Pareto-optimal solutions , 2003 .

[3]  Shigeyoshi Tsutsui,et al.  Advances in evolutionary computing: theory and applications , 2003 .

[4]  Kalyanmoy Deb,et al.  Simulated Binary Crossover for Continuous Search Space , 1995, Complex Syst..

[5]  J. Branke,et al.  Guidance in evolutionary multi-objective optimization , 2001 .

[6]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[7]  Peter J. Fleming,et al.  Genetic Algorithms for Multiobjective Optimization: FormulationDiscussion and Generalization , 1993, ICGA.

[8]  Lothar Thiele,et al.  Comparison of Multiobjective Evolutionary Algorithms: Empirical Results , 2000, Evolutionary Computation.

[9]  Kalyanmoy Deb,et al.  Multi-objective optimization using evolutionary algorithms , 2001, Wiley-Interscience series in systems and optimization.

[10]  Carlos A. Coello Coello,et al.  Handling preferences in evolutionary multiobjective optimization: a survey , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[11]  Marco Laumanns,et al.  Scalable multi-objective optimization test problems , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[12]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems , 2002, Genetic Algorithms and Evolutionary Computation.

[13]  Kalyanmoy Deb,et al.  A flexible optimization procedure for mechanical component design based on genetic adaptive search , 1998 .

[14]  Bernhard Sendhoff,et al.  Adapting Weighted Aggregation for Multiobjective Evolution Strategies , 2001, EMO.

[15]  Ian C. Parmee,et al.  Preferences and their application in evolutionary multiobjective optimization , 2002, IEEE Trans. Evol. Comput..

[16]  Kalyanmoy Deb,et al.  Multi-objective Genetic Algorithms: Problem Difficulties and Construction of Test Problems , 1999, Evolutionary Computation.

[17]  K. Deb.  Solving goal programming problems using multi-objective genetic algorithms , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).