An Intelligent Scatter with Estimation of Distribution for Tabu Search
This paper proposes an efficient optimization algorithm based on tabu search and estimation of distribution algorithms. The proposed algorithm estimates characteristics of distribution of solutions after performing tabu searches based on a marginal product model to acquire linkage information. Once correct linkage information is obtained, we can perform crossovers effectively without disrupting building blocks using the information. The proposed algorithm is expected to adapt to both global and local characteristics of solution space. Through empirical studies, we show the effectiveness of our approach.
Keywordstabu search extended compact genetic algorithm estimation of distribution algorithms
Unable to display preview. Download preview PDF.
- 2.Harik, G.: Linkage Learning via Probabilistic Modeling in the ECGA. IlliGAL Report No. 99010, University of Illinois at Urbana-Champaign, Urbana, IL (1999)Google Scholar