作者: Yukio Nakajima , Akihiko Abe
DOI: 10.1007/BF03167375
关键词:
摘要: A simple genetic algorithms (GAs) has been applied to generate the optimum pitch sequence. Though a GAs worked properly, there was problem of premature convergence. To solve this problem, we introduced new operator named growth and combined it with GAs. The operator, which is kind hill-climbing technique, function get local in small CPU time. GA generated better sequence than verified not have convergence even smaller population size. by improved noise performance such as pass-by compared current