Book Chapter10.1007/978-3-319-13356-0_35
A Novel Evolutionary Multi-objective Algorithm Based on S Metric Selection and M2M Population Decomposition
Lei Chen,Hai-Lin Liu,Chuan Lu,Chuan Lu,Yiu-ming Cheung,Jun Zhang +5 more
- 01 Jan 2015
- pp 441-452
6
TL;DR: A novel S metric selection evolutionary algorithm based on the population decomposition strategy MOEA/D-M2M is proposed to give a simple but effective method to improve the effectiveness of SMS based algorithm.
read more
Abstract: The excellent performance of evolutionary multi-objective algorithms based on S metric selection (SMS) has been identified by many researchers. However, huge computational effort of S metric calculation has limited the full application of those algorithms. This paper proposes a novel S metric selection evolutionary algorithm (SMS-M2M) based on the population decomposition strategy MOEA/D-M2M. In SMS-M2M, SMS is conducted in each subpopulation instead of the whole population, which can avoid the S metric calculation of the total population. The purpose of population decomposition is to directly reduce the huge computational effort of calculating S metric and thus to give a simple but effective method to improve the effectiveness of SMS based algorithm. SMS-M2M utilizes the same SMS with a popular SMS based evolutionary algorithm SMS-EMOA. Numerical studies of SMS-M2M and SMS-EMOA have shown that the M2M population decomposition can effectively reduce the computational effort of SMS, meanwhile the theoretic analysis identifies the efficiency and effectiveness of SMS-M2M.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
A Reference Vector Guided Evolutionary Algorithm for Many-Objective Optimization
TL;DR: In the proposed algorithm, a scalarization approach, termed angle-penalized distance, is adopted to balance convergence and diversity of the solutions in the high-dimensional objective space, and reference vectors are effective and cost-efficient for preference articulation, which is particularly desirable for many-objective optimization.
1.5K
Preference-Inspired Co-Evolutionary Algorithms With Local PCA Oriented Goal Vectors for Many-Objective Optimization
Zhe Shu,Weiping Wang +1 more
TL;DR: An improved PICEA-g is proposed to further resolve this long-standing problem and could have good performance on most of the test suites assessed in this research, and it performs very well compared with other many-objective optimization algorithms.
6
Combining modified inverted generational distance indicator with reference-vector-guided selection for many-objective optimization
TL;DR: A two-stage selection mechanism uses the IGD^+ indicator and the reference vector guidance method to select two sub-populations, which form the parent population at the next generation when solving MaOPs.
6
Preference-inspired coevolutionary algorithm with sparse autoencoder for many-objective optimization
Wei Wang,Shanxin Zhang,Wei-an Song,Wenlong Ge +3 more
TL;DR: The preference-inspired coevolutionary algorithm with sparse autoencoder (PICEA-g/SAE) is evaluated by three widely used benchmark suites and compared with nine classic multi-objective evolutionary algorithms to prove the advantages of the sparse aut Koencoder framework.
2
Adaptive stochastic fractal search algorithm for multi-objective optimization
HongShang Xu,Bei Dong,Xiaochang Liu,Ming Lei,Xiaojun Wu +4 more
TL;DR: This paper proposes RVSFS, a reference vector-guided stochastic fractal search algorithm for multi-objective optimization, which balances exploration and exploitation using adaptive mechanisms, Gaussian walks, and Levy flights, outperforming five state-of-the-art algorithms on 45 benchmarks.
1
References
•Book
Multi-Objective Optimization Using Evolutionary Algorithms
Kalyanmoy Deb,Deb Kalyanmoy +1 more
- 01 Jan 2001
TL;DR: This text provides an excellent introduction to the use of evolutionary algorithms in multi-objective optimization, allowing use as a graduate course text or for self-study.
•Book
Evolutionary Algorithms for Multiobjective Optimization: Methods and Applications
Eckart Zitzler
- 27 Dec 1999
TL;DR: The basic principles of evolutionary multiobjective optimization are discussed from an algorithm design perspective and the focus is on the major issues such as fitness assignment, diversity preservation, and elitism in general rather than on particular algorithms.
2.3K
SMS-EMOA : Multiobjective selection based on dominated hypervolume
TL;DR: A steady-state EMOA is proposed that features a selection operator based on the hypervolume measure combined with the concept of non-dominated sorting, thereby focussing on interesting regions of the Pareto front.
2K
A review of multiobjective test problems and a scalable test problem toolkit
TL;DR: This paper systematically review and analyze many problems from the EA literature, each belonging to the important class of real-valued, unconstrained, multiobjective test problems, and presents a flexible toolkit for constructing well-designed test problems.
A faster algorithm for calculating hypervolume
TL;DR: An algorithm for calculating hypervolume exactly, the Hypervolume by Slicing Objectives (HSO) algorithm, that is faster than any that has previously been published and increases the utility of hypervolume, both as a metric for general optimization algorithms and as a diversity mechanism for evolutionary algorithms.