Preference-Based Solution Selection Algorithm for Evolutionary Multiobjective Optimization

Kim, Jong-Hwan; Han, Ji-Hyeong; Kim, Ye-Hoon; Choi, Seung-Hwan; Kim, Eun-Soo

doi:10.1109/tevc.2010.2098412

Cited by 96 publications

(49 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To this end, statistical machine techniques, such as feature selection [36], principal component analysis (PCA) [37], [38], and maximum variance unfolding (MVU) [39] can be employed for objective reduction. In practice, users are often interested in only a part of the Pareto optimal solutions [40]. Therefore, if user preferences are available, preference-based approaches can be designed [41], [34], [42], [43].…”

mentioning

confidence: 99%

Diversity Assessment in Many-Objective Optimization

Wang

Jin

Yao

2017

IEEE Trans. Cybern.

206

View full text Add to dashboard Cite

Abstract-Maintaining diversity is one important aim of multiobjective optimization. However, diversity for many-objective optimization problems is less straightforward to define than for multi-objective optimization problems. Inspired by measures for biodiversity, we propose a new diversity metric for manyobjective optimization, which is an accumulation of the dissimilarity in the population, where an Lp-norm-based (p < 1) distance is adopted to measure the dissimilarity of solutions. Empirical results demonstrate our proposed metric can more accurately assess the diversity of solutions in various situations. We compare the diversity of the solutions obtained by four popular many-objective evolutionary algorithms using the proposed diversity metric on a large number of benchmark problems with two to ten objectives. The behaviors of different diversity maintenance methodologies in those algorithms are discussed in depth based on the experimental results. Finally, we show that the proposed diversity measure can also be employed for enhancing diversity maintenance or reference set generation in many-objective optimization.

show abstract

mentioning

confidence: 99%

Diversity Assessment in Many-Objective Optimization

Wang

Jin

Yao

2017

IEEE Trans. Cybern.

206

View full text Add to dashboard Cite

show abstract

“…There is no dispute over the point that there are reported applications in the literature, for example see Zhou et al (2011). However the scale of such problems pales in comparison to the size of problems that interior point methods can address, to the extent that Michalewicz (2012) million decision variables, constraints and objectives, while their convergence rate remains almost unaffected (Gondzio 2012), while to this day evolutionary algorithms mostly deal with problems with 2 or 3 objectives with relatively few addressing more, but no more than approximately 10, and with fewer than 100 decision variables (Kim et al 2012, Chiong and Kirley 2012, de Lange et al 2012, Wei et al 2012). Moreover there is not a single study available 1 that establishes a strong positive correlation between superior algorithm performance on test problems and real-world (and real-scale) problems.…”

Section: Discussionmentioning

confidence: 99%

An overview of population-based algorithms for multi-objective optimisation

Giagkiozis

Purshouse

Fleming

2013

International Journal of Systems Science

View full text Add to dashboard Cite

In this work we present an overview of the most prominent population-based algorithms and the methodologies used to extend them to multiple objective problems. Although not exact in the mathematical sense, it has long been recognized that population-based multi-objective optimization techniques for real-world applications are immensely valuable and versatile. These techniques are usually employed when exact optimization methods are not easily applicable or simply when, due to sheer complexity, such techniques could potentially be very costly. Another advantage is that since a population of decision vectors is considered in each generation these algorithms are implicitly parallelizable and can generate an approximation of the entire Pareto front (PF) at each iteration. A critique of their capabilities is also provided.

show abstract

“…The optimal solution corresponds to the one with the best partial evaluations. Despite not considering the criteria preferences, this kind of technique can be extended to include the degree of consideration for objectives [Kim et al, 2012]. MOP techniques, by expressing the objectives in functions where criteria are correlated, share the same issues as linear and integer programming approaches.…”

Section: One Of the Mathematical Optimization Methods Commonly Employmentioning

confidence: 99%