Rotated test problems for assessing the performance of multi-objective optimization algorithms

This paper presents four rotatable multi-objective test problems that are designed for testing EMO (Evolutionary Multi-objective Optimization) algorithms on their ability in dealing with parameter interactions. Such problems can be solved efficiently only through simultaneous improvements to each decision variable. Evaluation of EMO algorithms with respect to this class of problem has relevance to real-world problems, which are seldom separable. However, many EMO test problems do not have this characteristic. The proposed set of test problems in this paper is intended to address this important requirement. The design principles of these test problems and a description of each new test problem are presented. Experimental results on these problems using a Differential Evolution Multi-objective Optimization algorithm are presented and contrasted with the Non-dominated Sorting Genetic Algorithm II (NSGA-II).

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

[2]  Marco Laumanns,et al.  Scalable Test Problems for Evolutionary Multiobjective Optimization , 2005, Evolutionary Multiobjective Optimization.

[3]  David W. Corne,et al.  Instance Generators and Test Suites for the Multiobjective Quadratic Assignment Problem , 2003, EMO.

[4]  R. Salomon.  Re-evaluating genetic algorithm performance under coordinate rotation of benchmark functions. A survey of some theoretical and practical aspects of genetic algorithms. , 1996, Bio Systems.

[5]  Xiaodong Li,et al.  Incorporating directional information within a differential evolution algorithm for multi-objective optimization , 2006, GECCO.

[6]  Joshua D. Knowles,et al.  ParEGO: a hybrid algorithm with on-line landscape approximation for expensive multiobjective optimization problems , 2006, IEEE Transactions on Evolutionary Computation.

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

[8]  Tea Robic.  Performance of DEMO on New Test Problems: A Comparison Study , 2005 .

[9]  Xiaodong Li,et al.  Solving Rotated Multi-objective Optimization Problems Using Differential Evolution , 2004, Australian Conference on Artificial Intelligence.

[10]  Bernhard Sendhoff,et al.  On Test Functions for Evolutionary Multi-objective Optimization , 2004, PPSN.

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

[12]  R. Storn,et al.  Differential Evolution , 2004 .

[13]  R. Lyndon While,et al.  A Scalable Multi-objective Test Problem Toolkit , 2005, EMO.