一种新的模糊神经网络删剪策略

A new fuzzy neural network reduction strategy

模糊规则的数量直接决定模糊神经网络结构的复杂度和效率.基于神经网络自构行学习(NNSCL)算法,用共轭剃度预条件正则方程算法求取删除隐层神经元后的剩余权值,得到改进的NNSCL-1算法.将此算法应用到模糊神经网络的规则推理层,可以极大地优化网络的规则及结构,并且结构优化后不需要重新训练也能保持网络的精确度和泛化能力.仿真结果显示了此算法的有效性和可行性.

The number of fuzzy rules directly determines the complexity and efficiency of the fuzzy neural network（FNN）. Based on the neural network self-configuring learning（NNSCL） algorithm, the NNSCL-I algorithm is obtained by using the conjugate gradient precondition normal equation （CGPCNE） algorithm to adjust the remaining weghts after pruning nuurons. The NNSCL-I algorithm is applied in the rule-reasoning layer of the FNN to simplify its rules and structure in a great extent and preserve a good level of accuracy and generalization ability without retraining after pruning. The simulation results demonstrate the effectiveness and the feasibility of the algorithm.