Solving Dynamic Multiobjective Problem via Autoencoding Evolutionary Search

Solving Dynamic Multiobjective Problem via Autoencoding Evolutionary Search
复制标题

通过自动编码进化搜索解决动态多目标问题

DOI:
10.1109/tcyb.2020.3017017
复制
发表时间:
2020-10
影响因子:
11.8
通讯作者:
Kay Chen Tan
Kay Chen Tan
中科院分区:
计算机科学1区
文献类型:
--
作者:
Liang Feng;Wei Zhou;Weichen Liu;Yew-Soon Ong;Kay Chen Tan

文献摘要

参考文献

被引文献

相似文献

动态多目标优化问题(DMOP)是指目标随时间变化的多目标优化问题。由于DMOP在现实中的广泛应用,近十年来,DMOP引起了人们的广泛关注。在这篇文章中,我们建议通过自动编码进化搜索来解决DMOP。特别是,为了跟踪一个给定的DMOP的动态变化,一个自动编码器推导出预测移动的帕累托最优解的基础上获得的动态发生之前的非支配的解决方案。该自动编码器可以很容易地集成到现有的多目标进化算法(EA)中,例如NSGA-II,MOEA/D等,解决DMOP。与现有的方法相比,所提出的预测方法拥有一个封闭的形式的解决方案,从而不会带来太大的计算负担,在迭代进化搜索过程中。此外,所提出的动态变化的预测是从沿着动态优化过程中发现的非支配解中自动学习的,这可以提供更准确的帕累托最优解预测。为了研究所提出的自动编码进化搜索解决DMOP的性能,进行了全面的实证研究,通过比较三个国家的最先进的预测为基础的动态多目标EA。在常用的DMOP基准上获得的结果证实了所提出的方法的有效性。
Dynamic multiobjective optimization problem (DMOP) denotes the multiobjective optimization problem, which contains objectives that may vary over time. Due to the widespread applications of DMOP existed in reality, DMOP has attracted much research attention in the last decade. In this article, we propose to solve DMOPs via an autoencoding evolutionary search. In particular, for tracking the dynamic changes of a given DMOP, an autoencoder is derived to predict the moving of the Pareto-optimal solutions based on the nondominated solutions obtained before the dynamic occurs. This autoencoder can be easily integrated into the existing multiobjective evolutionary algorithms (EAs), for example, NSGA-II, MOEA/D, etc., for solving DMOP. In contrast to the existing approaches, the proposed prediction method holds a closed-form solution, which thus will not bring much computational burden in the iterative evolutionary search process. Furthermore, the proposed prediction of dynamic change is automatically learned from the nondominated solutions found along the dynamic optimization process, which could provide more accurate Pareto-optimal solution prediction. To investigate the performance of the proposed autoencoding evolutionary search for solving DMOP, comprehensive empirical studies have been conducted by comparing three state-of-the-art prediction-based dynamic multiobjective EAs. The results obtained on the commonly used DMOP benchmarks confirmed the efficacy of the proposed method.
DOI: 10.1109/tevc.2004.831456
发表时间: 2004-10-01
影响因子: 14.3
作者:
Farina, M;Deb, K;Amato, P
通讯作者: Amato, P
DOI: 10.1109/tevc.2008.920671
发表时间: 2009-02-01
影响因子: 14.3
作者:
Goh, Chi-Keong;Tan, Kay Chen
通讯作者: Tan, Kay Chen
DOI: --
发表时间: 1987-10
期刊: --
影响因子: --
作者:
D. Goldberg;R. Smith
通讯作者: D. Goldberg;R. Smith
DOI: 10.5555/1756006.1953039
发表时间: 2010-03
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
Pascal Vincent;H. Larochelle;Isabelle Lajoie;Yoshua Bengio;Pierre-Antoine Manzagol
通讯作者: Pascal Vincent;H. Larochelle;Isabelle Lajoie;Yoshua Bengio;Pierre-Antoine Manzagol
DOI: --
发表时间: 1993-06
期刊: --
影响因子: --
作者:
C. Fonseca;P. Fleming
通讯作者: C. Fonseca;P. Fleming