Decomposition-Based Interactive Evolutionary Algorithm for Multiple Objective Optimization

被引:31
作者
Tomczyk, Michal K. [1 ]
Kadzinski, Milosz [1 ]
机构
[1] Poznan Univ Tech, Inst Comp Sci, PL-60965 Poznan, Poland
关键词
Sociology; Optimization; Evolutionary computation; Analytical models; Additives; Monte Carlo methods; Decomposition; indirect preference information; interactive evolutionary hybrid; Monte Carlo (MC) simulation; multiple objective optimization (MOO); MULTIOBJECTIVE OPTIMIZATION; GENETIC ALGORITHM; RANKING;
D O I
10.1109/TEVC.2019.2915767
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We propose a decomposition-based interactive evolutionary algorithm (EA) for multiple objective optimization. During an evolutionary search, a decision maker (DM) is asked to compare pairwise solutions from the current population. Using the Monte Carlo simulation, the proposed algorithm generates from a uniform distribution a set of instances of the preference model compatible with such an indirect preference information. These instances are incorporated as the search directions with the aim of systematically converging a population toward the DMs most preferred region of the Pareto front. The experimental comparison proves that the proposed decomposition-based method outperforms the state-of-the-art interactive counterparts of the dominance-based EAs. We also show that the quality of constructed solutions is highly affected by the form of the incorporated preference model.
引用
收藏
页码:320 / 334
页数:15
相关论文
共 42 条
[1]  
Athanassopoulos AD, 1997, J OPER RES SOC, V48, P142, DOI 10.1057/palgrave.jors.2600345
[2]   Brain-Computer Evolutionary Multiobjective Optimization: A Genetic Algorithm Adapting to the Decision Maker [J].
Battiti, Roberto ;
Passerini, Andrea .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2010, 14 (05) :671-687
[3]   Comparative analysis of UTA multicriteria methods [J].
Beuthe, M ;
Scannella, G .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2001, 130 (02) :246-262
[4]  
Branke J, 2008, Multiobjective Optimization: Interactive and Evolutionary Approaches
[5]   Using Choquet integral as preference model in interactive evolutionary multiobjective optimization [J].
Branke, Juergen ;
Corrente, Salvatore ;
Greco, Salvatore ;
Slowinski, Roman ;
Zielniewicz, Piotr .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2016, 250 (03) :884-901
[6]   Learning Value Functions in Interactive Evolutionary Multiobjective Optimization [J].
Branke, Juergen ;
Greco, Salvatore ;
Slowinski, Roman ;
Zielniewicz, Piotr .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2015, 19 (01) :88-102
[7]   Heuristics for prioritizing pair-wise elicitation questions with additive multi-attribute value models [J].
Ciomek, Krzysztof ;
Kadzinski, Milosz ;
Tervonen, Tommi .
OMEGA-INTERNATIONAL JOURNAL OF MANAGEMENT SCIENCE, 2017, 71 :27-45
[8]   A fast and elitist multiobjective genetic algorithm: NSGA-II [J].
Deb, K ;
Pratap, A ;
Agarwal, S ;
Meyarivan, T .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (02) :182-197
[9]  
Deb K., 2003, DISTRIB COMPUT, P534
[10]  
Deb K., 2005, EVOL MULT OPT LONDON, DOI 10.1007/1-84628-137-76