Surrogate model assisted multi-objective differential evolution algorithm for performance optimization at software architecture level

Du XIN, Ni YOUCONG*, Wu XIAOBIN, Ye PENG, Xin YAO

*Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

1 Citation (Scopus)

Abstract

This paper proposes a surrogate model assisted differential evolutionary algorithm for performance optimization at the software architecture (SA) level, which is named SMDE4PO. In SMDE4PO, different strategies of crossover and mutation are adopted to enhance the algorithm’s search capability and speed up its convergence. Random forests are used as surrogate models to reduce the time of performance evaluation (i.e., fitness evaluation). Our comparative experiments on four different sizes of cases between SMDE4PO and NSGA-II are conducted. From the results, we can conclude that (1) SMDE4PO is significantly better than NSGA-II according to the three quality indicators of Contribution, Generation Distance and Hyper Volume; (2) By using random forests as surrogates, the run time of SMDE4PO is reduced by up to 48% in comparison with NSGA-II in our experiments. © Springer International Publishing AG 2017.
Original languageEnglish
Title of host publicationSimulated Evolution and Learning : 11th International Conference, SEAL 2017, Shenzhen, China, November 10–13, 2017, Proceedings
EditorsYuhui SHI, Kay Chen TAN, Mengjie ZHANG, Ke TANG, Xiaodong LI, Qingfu ZHANG, Ying TAN, Martin MIDDENDORF, Yaochu JIN
PublisherSpringer
Pages334-346
Number of pages13
ISBN (Electronic)9783319687599
ISBN (Print)9783319687582
DOIs
Publication statusPublished - 2017
Externally publishedYes
Event11th International Conference on Simulated Evolution and Learning, SEAL 2017 - Shenzhen, China
Duration: 10 Nov 201713 Nov 2017

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume10593
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameTheoretical Computer Science and General Issues
PublisherSpringer
ISSN (Print)2512-2010
ISSN (Electronic)2512-2029

Conference

Conference11th International Conference on Simulated Evolution and Learning, SEAL 2017
Abbreviated titleSEAL 2017
Country/TerritoryChina
CityShenzhen
Period10/11/1713/11/17

Keywords

  • Differential evolution
  • Performance optimization
  • Software architecture
  • Surrogate model

Fingerprint

Dive into the research topics of 'Surrogate model assisted multi-objective differential evolution algorithm for performance optimization at software architecture level'. Together they form a unique fingerprint.

Cite this