Dynamic Control of Genetic Algorithms in a Noisy Environment

In this paper, we present e cient algorithms for adjusting con guration parameters of genetic algorithms that operate in a noisy environment. Assuming that the population size is given, we address two problems speci cally important in a noisy environment. First, we study the duration-sizing problem that determines dynamically the duration of each generation. Next, we study the sample-allocation sizing problem that determines adaptively the number of evaluations taken from each population in a generation. For these two problems, we model the search process as a statistical selection process and derive equations useful for controlling the duration and the sample sizes. Our result shows that these adaptive procedures improve the performance of genetic algorithms over those of commonly used static ones.

• Type: Conference
• Year: 1993
• Where: ICGA
• Authors: Akiko N. Aizawa, Benjamin W. Wah
• Keywords: Genetic Algorithms, ICGA 1993, ICGA 1997, Noisy Environment, Sample-allocation Sizing Problem