在分佈估測演算法使用基因關聯之研究

Using Genewise Association in Estimation of Distribution Algorithm

馮笙瑀(Sheng-Yu Feng)

指導教授 ： 丁川康

國立中正大學/工學院/資訊工程學系/碩士(2015年)

摘要

無資料

關鍵字

在分佈估測演算法 ； 基因聯結

並列摘要

Estimation of distribution algorithm (EDA) is one of famous evolutionary algorithms (EAs) and has successfully solved various problems. However, EDA usually needs large population; specifically, the population size increase exponentially with the degree of interactions among sub-functions. We use a new statistical test approach, called FastANOVA, to find the linkage relation with small sample size. In this study, we conduct experiments to examine the linkage accuracy using trap-k problem and examine the usability by CEC2014 benchmark and uses NK landscape problem to examine the advantage of using FastANOVA in EDA. The experimental results show the FastANOVA improves the performance of EDA using small population.

並列關鍵字

Estimation of Distribution Algorithm ； Genewise Association

參考文獻

[6] R. A. Fisher. The Design of Experiments. Macmillan Pub Co, 1935.

[7] G. Harik. Linkage learning via probabilistic modeling in the ecga. Urbana, 51(61):801, 1999.

[8] G. Harik, F. G. Lobo, and D. E. Goldberg. The compact genetic algorithm. IEEE Transactions on Evolutionary Computation, 3(4):287-297, 1999.

[9] M. Hauschild and M. Pelikan. An introduction and survey of estimation of distribution algorithms. Swarm and Evolutionary Computation, 1(3):111-128, 2011.

[12] S. A. Kauman and E. D. Weinberger. The nk model of rugged fittness landscapes and its application to maturation of the immune response. Journal of Theoretical Biology, 141(2):211-245, 1989.