On the Evolutionary Optimization of Many Conflicting Objectives

By Robin Charles Purshouse, Peter John Fleming. IEEE Transactions on Evolutionary Computation, 2007

467 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

View Paper: https://www.semanticscholar.org/paper/120eb5ab32c6ef141b24e653b241429e3126d348

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

1. Yaochu Jin
2. Ran Cheng
3. Xing-yi Zhang
4. Ye Tian
5. Gary G. Yen
6. Yanan Sun
7. Hisao Ishibuchi
8. Xin Yao

In Computational Theory and Mathematics

1. Yaochu Jin
2. Xing-yi Zhang
3. Ye Tian
4. Ran Cheng
5. Gary G. Yen
6. Yanan Sun
7. Xin Yao
8. Shengxiang Yang

Citations in the corpus

• A survey on multi-objective evolutionary algorithms for many-objective problems (2014, methodology): “[57] results, in [83,84] it is shown that the NSGA-II performs very well with a small number of objectives but increasingly worse as the problem scales in the number of objectives.”
• A Grid-Based Evolutionary Algorithm for Many-Objective Optimization (2013, unknown): “, niche, crowding distance, clust ering, andk-th nearest distance [12]) not only cannot strengthen the selection pressure towards the Pareto front, but also hinde r the evolutionary search to some extent due to their favour on dom inance resistant solutions [56].”
• An Indicator-Based Multiobjective Evolutionary Algorithm With Reference Point Adaptation for Better Versatility (2018, background): “objectives, their performance deteriorates rapidly as the number of objectives increases on MaOPs, mainly due to the phenomenon known as dominance resistance [6].”
• Overcoming the bootstrap problem in evolutionary robotics using behavioral diversity (2009, background): “The previously described approach requires to optimize many objectives whereas current multiobjective evolutionary algorithms have poor performance when more than three objectives are used [27], [28].”
• Evolutionary Constrained Multiobjective Optimization: Test Suite Construction and Performance Comparisons (2019, background): “As pointed out in [51], the proportion of nondominated solutions in the population grows exponentially with the increase of the number of objectives.”
• A Vector Angle-Based Evolutionary Algorithm for Unconstrained Many-Objective Optimization (2017, background): “It has been shown that Pareto-based MOEAs, such as nondominated sorting genetic algorithm II (NSGA-II) [17], the improved strength Pareto evolutionary algorithm (SPEA2) [19], the Pareto archived evolution strategy [20], and Pareto envelope-based selection algorithm II (PESAII) [21] encountered diffi”
• Novelty-Based Multiobjectivization (2011, unknown): “However, MOEA give better results with two objectives than three [22], [23]; consequently this variant may encounter some difficulties.”
• Shift-Based Density Estimation for Pareto-Based Algorithms in Many-Objective Optimization (2014, background): “Most current EMO algorithms, which work well on problems with two or three objectives, noticeably deteriorate their search ability when more objectives are involved [46], [48], [65].”
• IGD Indicator-Based Evolutionary Algorithm for Many-Objective Optimization Problems (2018, background): “selected in a large search space is not necessary to generate promising offspring [7], [60].”
• A Decision Variable Clustering-Based Evolutionary Algorithm for Large-Scale Many-Objective Optimization (2018, background): “rithms (MOEAs) designed for dealing with traditional multiobjective optimization problems (MOPs) which involve only two or three objectives due to two major issues [5]–[9].”
• Evolutionary algorithms and other metaheuristics in water resources: Current status, research challenges and future directions (2014, background): “Instead, they are best solved using multiple performance criteria (referred to as “many-objective” problems (Purshouse and Fleming, 2007)) that capture a broad range of system properties and stakeholder preferences, ultimately requiring that the decision maker choose from a suite of Pareto optimal d”
• A historical review of evolutionary learning methods for Mamdani-type fuzzy rule-based systems: Designing interpretable genetic fuzzy systems (2011, unknown): “Nevertheless, recent proposals in the EMO field have managed to deal with a significantly large number of objectives inwhat is called evolutionarymany-objective optimization [98,137] and their adaptation to multiobjective GFS has become feasible.”
• A New Dominance Relation-Based Evolutionary Algorithm for Many-Objective Optimization (2016, background): “This is mainly because, in highdimensional objective space, there is a greater probability that solutions widely distant from each other will be selected to recombine and generate poorer performance offspring solutions, known as lethals [26], [28].”
• Preference-Inspired Coevolutionary Algorithms for Many-Objective Optimization (2013, methodology): “Particularly, the “sweetspot” [7] of algorithm parameter settings that yield good performance may greatly contract [5], [6], i.”
• Many-Objective Evolutionary Algorithms (2015, background): “This is mainly due to the DR phenomenon and the active diversity promotion (ADP) mechanisms [Purshouse and Fleming 2007].”
• Borg: An Auto-Adaptive Many-Objective Evolutionary Computing Framework (2013, background): “This phenomenon is termed active diversity maintenance (Purshouse and Fleming, 2007).”
• A Survey of Multiobjective Evolutionary Algorithms Based on Decomposition (2017, background): “Since the last decade, a considerable amount of effort has been spent by researchers in developing efficient algorithms for solving MaOPs [12], [112]–[121].”