Theoretical and Empirical Investigations on Difficulty in Structure Learning by Estimation of Distribution Algorithms
Theoretical and Empirical Investigations on Difficulty in Structure Learning by Estimation of Distribution Algorithms
复制标题
DOI：
10.1109/icsmc.2006.384384
复制
发表时间：
2006-10
期刊：
影响因子：
--
通讯作者：
Miwako Tsuji;M. Munetomo;K. Akama
中科院分区：
文献类型：
--
作者：
Miwako Tsuji;M. Munetomo;K. Akama
分布估计算法（EDA）是从遗传算法（GAs）发展而来的基于种群的进化算法. EDA建立有希望的解决方案的概率模型，以指导搜索空间的进一步探索。它们被认为与GA的行为方式相似。在本文中，我们通过设计一个EDA困难函数，其中与问题结构不一致的模式有时会压倒与问题结构一致的模式，展示了它们在EDA应用中的不同行为和困难。
Estimation of distribution algorithms (EDAs) are population based evolutionary algorithms derived from genetic algorithms (GAs) . EDAs build probabilistic models of promising solutions to guide further exploration of the search space. They have been considered to behave in similar way to GAs. In this paper, we show their different behaviors and difficulties in applications of EDAs by designing an EDA difficult function in which schemata that are not consistent with problem structure sometimes overwhelm those that are.