Show simple item record

dc.contributor.authorWu, Xiaofang
dc.contributor.authorLi, Changhe
dc.contributor.authorZeng, Sanyou
dc.contributor.authorYang, Shengxiang
dc.date.accessioned2019-11-25T13:22:44Z
dc.date.available2019-11-25T13:22:44Z
dc.date.issued2019-11
dc.identifier.citationWu, X., Li, C., Zeng, S. and Yang, S. (2019) A novel multi-objective evolutionary algorithm based on space partitioning. Proceedings of the 11th International Symposium on Intelligence Computation and Applications,en
dc.identifier.urihttps://dora.dmu.ac.uk/handle/2086/18864
dc.description.abstractTo design an e ective multi-objective optimization evolutionary algorithms (MOEA), we need to address the following issues: 1) the sensitivity to the shape of true Pareto front (PF) on decomposition-based MOEAs; 2) the loss of diversity due to paying so much attention to the convergence on domination-based MOEAs; 3) the curse of dimensionality for many-objective optimization problems on grid-based MOEAs. This paper proposes an MOEA based on space partitioning (MOEA-SP) to address the above issues. In MOEA-SP, subspaces, partitioned by a k-dimensional tree (kd-tree), are sorted according to a bi-indicator criterion de ned in this paper. Subspace-oriented and Max-Min selection methods are introduced to increase selection pressure and maintain diversity, respectively. Experimental studies show that MOEA-SP outperforms several compared algorithms on a set of benchmarks.en
dc.language.isoen_USen
dc.publisherSpringeren
dc.subjectMulti-objective optimizationen
dc.subjectkd-tree space partitioningen
dc.subjectMax-Min methoden
dc.titleA novel multi-objective evolutionary algorithm based on space partitioningen
dc.typeConferenceen
dc.peerreviewedYesen
dc.funderOther external funder (please detail below)en
dc.projectid61673355en
dc.cclicenceCC-BY-NC-NDen
dc.date.acceptance2019-10-01
dc.researchinstituteInstitute of Artificial Intelligence (IAI)en
dc.funder.otherNational Natural Science Foundation of Chinaen


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record