论文部分内容阅读
本文在分析了密歇根方法和匹茨堡方法的基础上,提出了一种新型分层协同进化学习方法。该方法由上述两类种群构成,此两类种群分别属于不同的智能层次,进行协同进化来实现智能,种群内部各自独立地采取不同的遗传操作,种群之间使用交互算法进行交流。实验表明:该方法能改善分类器系统的短视特性并提高其智能。
Based on the analysis of the Michigan method and the Pittsburgh method, this paper proposes a new hierarchical cooperative evolution learning method. The method is composed of the above two types of populations, which belong to different intelligent levels respectively, and carry out cooperative evolution to realize intelligence. Different genetic operations are taken independently inside the populations, and the interaction between the populations is implemented using an interactive algorithm. Experiments show that this method can improve the short-sightedness of the classifier system and enhance its intelligence.