Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms

This work introduces a hybrid between an elitist multi-objective evolutionary algorithm and a gradient-based descent method, which is applied only to certain (selected) solutions. Our proposed approach requires a low number of objective function evaluations to converge to a few points in the Pareto front. Then, the rest of the Pareto front is reconstructed using a method based on rough sets theory, which also requires a low number of objective function evaluations. Emphasis is placed on the effectiveness of our proposed hybrid approach when increasing the number of decision variables, and a study of the scalability of our approach is also presented.

[1]  Jörg Fliege,et al.  Steepest descent methods for multicriteria optimization , 2000, Math. Methods Oper. Res..

[2]  Joshua D. Knowles,et al.  Memetic Algorithms for Multiobjective Optimization: Issues, Methods and Prospects , 2004 .

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

[4]  Griewank,et al.  On automatic differentiation , 1988 .

[5]  Martin Brown,et al.  Effective Use of Directional Information in Multi-objective Evolutionary Computation , 2003, GECCO.

[6]  Kalyanmoy Deb,et al.  A Local Search Based Evolutionary Multi-objective Optimization Approach for Fast and Accurate Convergence , 2008, PPSN.

[7]  Peter A. N. Bosman,et al.  Combining gradient techniques for numerical multi-objective evolutionary optimization , 2006, GECCO '06.

[8]  Enrique Alba,et al.  A comparative study of the effect of parameter scalability in multi-objective metaheuristics , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[9]  X. Yi.  On Automatic Differentiation , 2005 .

[10]  S. Schäffler,et al.  Stochastic Method for the Solution of Unconstrained Vector Optimization Problems , 2002 .

[11]  Isao Ono,et al.  Uniform sampling of local pareto-optimal solution curves by pareto path following and its applications in multi-objective GA , 2007, GECCO '07.

[12]  C. Coello,et al.  Multiobjective optimization using a micro-genetic algorithm , 2001 .

[13]  Shigenobu Kobayashi,et al.  Hybridization of genetic algorithm and local search in multiobjective function optimization: recommendation of GA then LS , 2006, GECCO '06.

[14]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems (Genetic and Evolutionary Computation) , 2006 .

[15]  Janusz Zalewski,et al.  Rough sets: Theoretical aspects of reasoning about data , 1996 .

[16]  Pradyumn Kumar Shukla,et al.  On Gradient Based Local Search Methods in Unconstrained Evolutionary Multi-objective Optimization , 2007, EMO.

[17]  Frank Neumann,et al.  Do additional objectives make a problem harder? , 2007, GECCO '07.

[18]  Peter A. N. Bosman,et al.  Exploiting gradient information in numerical multi--objective evolutionary optimization , 2005, GECCO '05.

[19]  Carlos A. Coello Coello,et al.  On the Use of Projected Gradients for Constrained Multiobjective Optimization Problems , 2008, PPSN.

[20]  Oliver Schütze,et al.  On Continuation Methods for the Numerical Treatment of Multi-Objective Optimization Problems , 2005, Practical Approaches to Multi-Objective Optimization.