Genetic algorithms with noisy fitness

被引：9
作者：
Zhai, W
Kelly, P
Gong, WB
机构：
[1] Dept. of Elec./Computer Engineering, University of Massachusetts, Amherst
来源：
MATHEMATICAL AND COMPUTER MODELLING | 1996年 / 23卷 / 11-12期
基金：
美国国家科学基金会;
关键词：
genetic algorithms; optimization; estimated performance;
D O I：
10.1016/0895-7177(96)00068-4
中图分类号：
TP39 [计算机的应用];
学科分类号：
081203 ; 0835 ;
摘要：
The convergence properties of genetic algorithms with noisy fitness information are studied here. In the proposed scheme, hypothesis testing methods are used to compare sample fitness values. The ''best'' individual of each generation is kept and a greater-than-zero mutation rate is used so that every individual will be generated with positive probability in each generation. The convergence criterion is different from the frequently-used uniform population criterion; instead, the sequence of the ''best'' individual in each generation is considered, and the algorithm is regarded as convergent if the sequence of the ''best'' individuals converges with probability one to a point with optimal average fitness.
引用
收藏
页码：131 / 142
页数：12
相关论文
共 14 条
  • [1] [Anonymous], THESIS U FLORIDA
  • [2] Fitzpatrick J. M., 1988, Machine Learning, V3, P101
  • [3] Gantmacher F.R., 1977, The Theory of Matrices
  • [4] GONG WB, 1993, PROCEEDINGS OF THE 32ND IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4, P2211, DOI 10.1109/CDC.1993.325592
  • [5] GONG WB, 1992, PROCEEDINGS OF THE 31ST IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4, P795, DOI 10.1109/CDC.1992.371616
  • [6] GONG WB, 1996, CCS96001 U MASS AMH
  • [7] GREFENSTETTE JJ, 1981, CS8119 VAND U COMP S
  • [8] GREFENSTETTE JJ, 1987, P 1 INT C GEN ALG TH, P112
  • [9] HO YC, 1992, J DEDS APR
  • [10] HOLLAND JH, 1975, ADAPTATION NATURAL A
12