[Updated on Oct. 4] Integration of CiNii Articles into CiNii Research from April 1, 2022

SSEの世代交代モデルを改良したcSSEについて

Bibliographic Information

Other Title
  • A Study on cSSE that Improves Generation Alternation Model of SSE

Abstract

組合せ最適化問題を効率良く解く進化的計算手法の1つに,確率的スキーマ貪欲法(Stochastic Schemata Exploiter: SSE)がある.SSEは,母集団に多様性を持たせることで,優れた収束特性を維持したまま大域的探索能力が向上する可能性がある.そこで,本論文ではSSEの世代交代モデルを改良したCross generational elitist selection SSE(cSSE)を提案する.さらに,0/1組合せ最適化問題において,SSE,cSSEをMinimal Generation Gap(MGG)に基づくGA,Bayesian Optimization Algorithm(BOA)と性能比較を行い,それらの探索性能を検討する.その結果,cSSEは優れた収束特性と大域的探索能力を有していることが分かった. The Stochastic Schemata Exploiter (SSE) is one of the evolutionary optimization algorithms for solving the combinatorial optimization problems. The SSE can improve the global search ability by maintaining the diversity of the population. In this paper, we present the Cross generational elitist selection SSE (cSSE) algorithms which improves the generation alternation model of the SSE. The SSE and the cSSE are compared with the GA with the Minimal Generation Gap (MGG) and the Bayesian Optimization Algorithm (BOA) in 0/1 combinatorial optimization problem in order to discuss their convergence property. As a result, we indicate that cSSE has an excellent convergence property and the global search ability.

identifier:http://hdl.handle.net/2237/10306

Journal

Citations (0)*help

See more

References(0)*help

See more

Related Articles

See more

Related Data

See more

Related Books

See more

Related Dissertations

See more

Related Projects

See more

Related Products

See more

Details

Back to top