On measuring multiobjective evolutionary algorithm performance

Veldhuizen, David A. Van; Lamont, Gary B.

doi:10.1109/cec.2000.870296

Cited by 315 publications

(181 citation statements)

References 11 publications

Supporting

Mentioning

174

Contrasting

Unclassified

Order By: Relevance

“…However, these studies mainly cover preferences on solutions and not preferences on sets. Furthermore, there is a large amount of publications that deal with the definition and the application of quantitative quality measures for Pareto set approximations [18], [34], [29], [20], [35]. These quality measures or quality indicators reflect set preferences and have been widely employed to compare the outcomes generated by different MOEAs.…”

Section: Introductionmentioning

confidence: 99%

On Set-Based Multiobjective Optimization

Zitzler

Thiele

Bader

2010

IEEE Trans. Evol. Computat.

156

104

View full text Add to dashboard Cite

Abstract-Assuming that evolutionary multiobjective optimization (EMO) mainly deals with set problems, one can identify three core questions in this area of research: (i) how to formalize what type of Pareto set approximation is sought, (ii) how to use this information within an algorithm to efficiently search for a good Pareto set approximation, and (iii) how to compare the Pareto set approximations generated by different optimizers with respect to the formalized optimization goal. There is a vast amount of studies addressing these issues from different angles, but so far only few studies can be found that consider all questions under one roof. This paper is an attempt to summarize recent developments in the EMO field within a unifying theory of set-based multiobjective search. It discusses how preference relations on sets can be formally defined, gives examples for selected user preferences, and proposes a general, preference-independent hill climber for multiobjective optimization with theoretical convergence properties. Furthermore, it shows how to use set preference relations for statistical performance assessment and provides corresponding experimental results. The proposed methodology brings together preference articulation, algorithm design, and performance assessment under one framework and thereby opens up a new perspective on EMO.

show abstract

Section: Introductionmentioning

confidence: 99%

On Set-Based Multiobjective Optimization

Zitzler

Thiele

Bader

2010

IEEE Trans. Evol. Computat.

156

104

View full text Add to dashboard Cite

show abstract

“…The solutions in the nondominated set may have a high diversity, however the ultimate goal of optimisation is to find solutions with optimal fitness. Unary fitness metrics found in the literature measure the distance of the nondominated solutions to the Pareto-optimal front, such as the proximity indicator [19] and the generation distance [20]. The optimal solutions often are not known, especially when solving previously unknown NP-hard problems.…”

Section: Performance Metrics For Multiobjective Optimisation Algormentioning

confidence: 99%

Studying feedback mechanisms for adaptive parameter control in evolutionary algorithms

Aleti

Moser

2013

2013 IEEE Congress on Evolutionary Computation

View full text Add to dashboard Cite

Abstract-The performance of an Evolutionary Algorithm (EA) is greatly affected by the settings of its strategy parameters. An effective solution to the parameterisation problem is adaptive parameter control, which applies learning methods that use feedback from the optimisation process to evaluate the effect of parameter value choices and adjust the parameter values over the iterations. At every iteration of an EA, the performance of an EA is reported and employed by the feedback mechanism as an indication of the success of the parameterisation of the algorithm instance.Many approaches to collect information about the algorithm's performance exist in singleobjective optimisation. In this work, we review the most recent and prominent approaches. In multiobjective optimisation, establishing a single scalar which can report the algorithm's performance as feedback for adaptive parameter control is a complex task. Existing performance measures of multiobjective optimisation are generally used as feedback for the optimisation process. We discuss the properties of these measures and present an empirical evaluation of the binary hypervolume and +-indicators as feedback for adaptive parameter control.

show abstract

“…This ensures that P true contains all of the D paths computed by Dijstra's algorithm on each of the D criteria and other pareto optimal solutions generated by the EA and the k-shortest path algorithm. The approach we use to generating P true is similar to that used in Okabe et al (2003) and Baran et al (2001) For a more indepth coverage of quality metrics for multicriteria optimization the work by Van Veldhuizen and Lamont (2000) provides examples of a wide range of metrics.…”

Section: Assessing the Quality Of The Ea Approachmentioning

confidence: 99%

An evolutionary algorithm for multicriteria path optimization problems

Mooney

Winstanley

2006

International Journal of Geographical Information Science

View full text Add to dashboard Cite

For many years researchers and decision makers (DMs) faced with multicriteria shortest path problems (MSPPs) have resorted to reductions to the classical shortest path problem (SPP) by means of weighted linear combinations of the criteria. Algorithmic and approximation schemes are available to solve MSPPs but these approaches often display complexities prohibitive to their implementation on real-world applications. This paper describes the development of an Evolutionary Algorithm (EA) approach to MSPPs on networks with multiple independent criteria. The EA approach is shown to sufficiently explore the underlying network space, generate large candidate path sets, and evolve high quality approximations to the optimal MSPP solution(s). Opportunities for early termination of the EA in time-critical applications are also offered. Among the issues for further work is the integration of the EA as a tool within a GIS for path optimization.

show abstract

On measuring multiobjective evolutionary algorithm performance

Cited by 315 publications

References 11 publications

On Set-Based Multiobjective Optimization

On Set-Based Multiobjective Optimization

Studying feedback mechanisms for adaptive parameter control in evolutionary algorithms

An evolutionary algorithm for multicriteria path optimization problems

Contact Info

Product

Resources

About