基于郭涛算法的演化神经网络

Evolutionary neural networks based on GT algorithm

提出了一种新的演化神经网络算法GTEANN,该算法基于高效的郭涛算法,同时完成在网络结构空间和权值空间的搜索,以实现前馈神经网络的自动化设计。本方法采用的编码方案直观有效,基于该编码表示,神经网络的学习过程是一个复杂的混合整实数非线性规划问题,例如杂交操作包括网络的同构和规整处理。初步实验结果表明该方法收敛,能够达到根据训练样本自动优化设计多层前馈神经网络的目的。

A new method of evolutionary neural networks,called evolutionary neural networks based GT algorithm （GTEANN）,is proposed in this study.ln this method,GT algorithm is used to simultaneousely search the satisfied structure and weights for feedforward neural networks.A straightforward effective encoding scheme for feedforward neural networks is adopted and only crossover operators are used with special topological isomorphism and regularization process.The learing process of network is a complex mixed-integer nonlinear optimization problem.The initial results of experiments indicate that GTEANN can automatically design and optimaize neural networks using the training sets.