SMS-EMOA: Multiobjective selection based on dominated hypervolume

By Nicola Beume, Boris Naujoks, Michael T. M. Emmerich. European Journal of Operational Research, 2007

2,020 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/88d5f76bcf9d4c4e888661cb12ffe7f8e669d310

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Management Science and Operations Research

Citations in the corpus

A Reference Vector Guided Evolutionary Algorithm for Many-Objective Optimization (2016, methodology): “, the indicator based evolutionary a lgorithm (IBEA) [42], the S-metric selection based evolutio nary multi-objective algorithm (SMS-EMOA) [43], a dynamic neighborhood multi-objective evolutionary algorithm bas ed on hypervolume indicator (DNMOEA/HI) [44], and the fast hypervolume based evolutionary”
An Evolutionary Algorithm for Large-Scale Sparse Multiobjective Optimization Problems (2020, methodology): “For the above sparse MOPs, most work solves them via existing MOEAs like NSGA-II [23], SMS-EMOA [25], and MOEA/D [46], and some work also enhances the performance by adopting customized search strategies.”
Test Problems for Large-Scale Multiobjective and Many-Objective Optimization (2017, methodology): “In general, multi objective evolutionary algorithms (MOEAs) can be categorized into three groups, including dominance based methods [6]–[8], decomposition based methods [9]–[11], and per formance indicator based methods [12], [13].”
A Survey of Multiobjective Evolutionary Algorithms for Data Mining: Part I (2014, background): “The main advantage of indicatorbased MOEAs such as SMS-EMOA is that they seem to scale better in the presence of many objectives (four or more).”
Stable Matching-Based Selection in Evolutionary Multiobjective Optimization (2014, neutral): “, [7], [8], [9]) and decomposition-based MOEAs (e.”
Multi-objective training of Generative Adversarial Networks with multiple discriminators (2019, background): “The hypervolume was originally introduced as a quantitative metric for coverage and convergence of Pareto-optimal fronts obtained through population-based algorithms (Beume et al., 2007).”
A Decision Variable Clustering-Based Evolutionary Algorithm for Large-Scale Many-Objective Optimization (2018, background): “Among many others [23]–[26], some representative approaches of this category are indicator-based evolutionary algorithm [27], S-metric selection based evolutionary multiobjective optimization algorithm [28], and hypervolume (HV)-based evolutionary algorithm [29].”
An easy-to-use real-world multi-objective optimization problem suite (2020, methodology): “A number of EMOAs have been proposed, including NSGA-II [2], SPEA2 [3], IBEA [4], SMS-EMOA [5], MOEA/D [6], and NSGA-III [7].”
Exploratory landscape analysis (2011, methodology): “We use an adapted SMS-EMOA, as proposed by Beume et al. (2007), for our dis crete parameter space.”
PlatEMO: A MATLAB Platform for Evolutionary Multi-Objective Optimization [Educational Forum] (2017, methodology): “Algorithm Year of Description Publication Multi-Objective Genetic Algorithms SPEA2 [14] 2001 Strength Pareto evolutionary algorithm 2 PSEA-II [15] 2001 Pareto envelope-based selection algorithm II NSGA-II [13] 2002 Non-dominated sorting genetic algorithm II ǫ-MOEA [28] 2003 Multi-objective evolution”
Bio-inspired computation: Where we stand and what's next (2019, methodology): “However, the mere use of a performance indicator in the density estimator was discovered to suffice for rendering a good performance (as in e.g., SMS-EMOA in which the hypervolume is adopted as a density estimator that replaces the crowded comparison operator of NSGAII).”
Multi-Objective Evolutionary Federated Learning (2018, methodology): “In order to monitor the convergence of the multi-objective optimization, the hypervolumes calculated based on the nondominated solution set in the population over the generations [39] in evolving MLP and CNN on non-IID datasets are plotted in Fig.”
A Method for a Posteriori Identification of Knee Points Based on Solution Density (2018, methodology): “MOEAs can be roughly classified into four categories, namely, dominance-based methods [1], [2], [3], [4], indicators or metrics based approaches [5], [6], [7], aggregation or decomposition based methods [8], [9], [10], [11], as well as preference-based methods [12], [13], [14], [15].”
Performance of Decomposition-Based Many-Objective Algorithms Strongly Depends on Pareto Front Shapes (2017, methodology): “One is an indicator-based approach such as S metric selection evolutionary multiobjective optimization algorithm (SMS-EMOA) [22] and hypervolume estimation algorithm for multiobjective optimization (HypE) [23].”