摘要： 为了更有效地优化前向神经网络的求解能力，提出了一种新的综合的转换函数，将多层感知机和RBF 神经网络更有机地结合起来，以产生灵活的决策边界。在此基础上推导出了相应的学习算法。并通过对实际的模式分类问题的仿真，将文中的方法与带动量项BP算法、CSFN、RBF 等算法进行了比较，验证了其有效性。

关键词: 转换函数；径向基函数；多层感知机；混合网络；学习算法

Abstract: In order to effectively optimizing the solution of feed-forward neural network, a new general transfer function is proposed that effectively unifies the inputs of multilayer perception and radial basis function to provide flexible decision border. A new algorithm based on gradient descent and error propagation is proposed. Several pattern classification example simulations are made to verify the validity of the proposed algorithm by comparing the proposed transfer function and learning algorithm with BP algorithm adding momentum term, CSFN and RBF

Key words: Transfer function; Radial basis function; Multilayer perception; Hybrid network; Learning algorithm