摘要
现有的评价启发式算法性能的方法有许多种,但是它们的评价标准各不相同。从启发式算法的共性出发,通过分析影响算法性能的各方面因素,提出评价启发式优化算法的一般性方法。介绍了为寻找最简洁版本优化算法而得到的学习算法(LA),并且将该方法运用到对学习算法性能的评价中,得出了学习算法是一种优于经典的遗传算法(GA)和微粒群算法(PSO)的有效的启发式优化算法。
There are many kinds of evaluation methods for heuristic algorithms, but their evaluation standards are different. Through analyzing the various factors affecting algorithm performance, this paper presented a general method to evaluate heuristic optimization algorithms based on their commonnesses. Learning Algorithm （LA） was introduced to find the most concise version of the optimization algorithm. The general evaluation method was also applied in the evaluation of LA. A conclusion is drawn that LA is a kind of effective heuristic optimization algorithm, and its optimized performance is superior to the classical algorithms, such as Genetic Algorithm （GA） and Particle Swarm Optimization （PSO）.
出处
《计算机应用》
CSCD
北大核心
2010年第A01期76-79,82,共5页
journal of Computer Applications
基金
国家自然科学基金青年基金资助项目(60801035)
关键词
启发式算法
性能评价
学习算法
遗传算法
微粒群算法
heuristic algorithm
performance evaluation
Learning Algorithm （LA）
Genetic Algorithm （GA）
Particle Swarm Optimization （PSO）
