作者: Martin Pelikan , David E. Goldberg , Kumara Sastry
DOI:
关键词: Evolutionary algorithm 、 Fitness approximation 、 Population size 、 Statistics 、 Mathematics 、 Fitness inheritance 、 Function (mathematics) 、 Inheritance (genetic algorithm) 、 Convergence (routing)
摘要: This paper studies fitness inheritance as an efficiency enhancement technique for genetic and evolutionary algorithms. Convergence population-sizing models are derived compared with experimental results. These optimized greatest speed-up the optimal proportion to obtain such a is derived. Results on OneMax problems show that when effects considered in model, number of function evaluations reduced by 20% use inheritance. indicate fixed population size, can be 70% using simple technique.