进化计算与人工神经网络的结合

进化计算和人工神经网络作为两个工具在众多的研究领域得到了广泛应用。进化计算和人工神经网络本身也得到了很大的发展。类似于生物神经网络，人工神经网络（ Ａ Ｎ Ｎ） 由一些简单的处理单元组成，这些单元通过权值的连接而相互作用。 Ａ Ｎ Ｎ 因其鲁棒性、并行性及学习能力受到特别的重视。进化计算体现了生物进化中的四个要素，即：繁殖、变异、竞争和自然选择。目前泛指各种基于生物进化原理的仿真计算方法的总称。文中首先介绍了进化计算的有关概念，包括遗传算法、进化策略等，其次就其与人工神经网络技术相结合的方法作了进一步分析探讨。主要集中于进化的网络结构设计、进化的网络训练及其它结合方法等方面的有关问题。

Combination of evolutionary computation and artificial neural network

As two kinds of tools, evolutionary computation (EC) and artificial neural network get wide applications in many research areas. Similar to biologic neural network, artificial neural network is composed of some simple processing units. These units interact through weight connection. Artificial neural network gets recognition for its robust, parallel and learning ability. Evolutionary computation shows four elements of biologic evolution: propagation, variation, competition and natural selection. At present, it refers to the general name of diversified simulation computation methods based on biologic evolution theory. In this paper, some concepts related to evolutionary computation are introduced firstly, including genetic algorithms, evolutionary strategy, evolutionary programming etc. Then the combination of EC and artificial neural network and some problems are discussed, especially on evolutionary artificial neural network design and evoluationary artificial neural network training, etc. ;