Multiobjective quantum-inspired evolutionary algorithm = 양자 개념을 도입한 다목적 진화 알고리즘

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 190
  • Download : 0
Evolutionary algorithms (EAs) inspired from the processes of evolution in nature are based on stochastic search mechanisms. In recent studies, estimation of distribution algorithms (EDAs) using probabilistic modeling have represented the EAs instead of conventional genetic algorithms (GAs), which use crossover and mutation. Quantum-inspired evolutionary algorithm (QEA), one of the novel EDAs, is proved to be better than conventional GAs for single-objective optimization problems. Current researches in EAs are focused on simultaneous optimization problems of several objectives, which are called multiobjective optimization problems (MOPs) and multiobjective evolutionary algorithms (MOEAs) allow to obtain a set of solutions. Since MOEAs provide a set of nondominated solutions, decision making of selecting a preferred one out of them is required in real application. However, there have been a few researches on MOEA in which the user`s preferences are incorporated for this purpose i.e. multiple criteria decision making (MCDM). This thesis proposes multiobjective quantum-inspired evolutionary algorithm (MQEA) which employs the concept and principles of quantum computing to improve the proximity to Pareto-optimal front and diversity of nondominated solutions for MOPs. MQEA employs the basic characteristic of EDAs such as offspring sampling by the probabilistic representation. It has an advantage to maintain an elitism by storing nondominated solutions of archive externally. Global random migration of nondominated solutions in the archive is adopted to preserve the diversity of solutions of each sub-population. Cooperative coevolutionary concept of MQEA based on parallel structure has robust search ability on more complex and higher-dimensional objective problems. Moreover, this thesis proposes preference-based solution selection algorithm (PSSA) by which user can select a desired one out of nondominated solutions obtained by any one of MOEAs. The PSSA, which is a ki...
Advisors
Kim, Jong-Hwanresearcher김종환researcher
Description
한국과학기술원 : 전기 및 전자공학과,
Publisher
한국과학기술원
Issue Date
2010
Identifier
418791/325007  / 020055025
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 전기 및 전자공학과, 2010.2, [ xi, 116 p. ]

Keywords

Preference-based solution selection; Robot soccer; Fuzzy path planning; Evolutionary multiobjective optimization; Multiobjective quantum-inspired evolutionary algorithm; 양자 개념을 도입한 다목적 진화 알고리즘; 선호도기반 해 선택; 로봇 축구; 퍼지 경로 계획; 다목적 진화 최적화

URI
http://hdl.handle.net/10203/35585
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=418791&flag=dissertation
Appears in Collection
EE-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0