基于伪并行混合遗传算法的神经网络优化

在分析并行多物种遗传算法应用于神经网络拓扑结构的设计和学习之后，提出一种伪并行遗传（PPGA-MBP）混合算法，结合改进的BP算法对多层前馈神经网络的拓扑结构进行优化。算法编码采用基于实数的层次混合方式，允许两个不同结构的网络个体交叉生成有效子个体。利用该算法对N-Parity问题进行了实验仿真，并对算法中评价函数各部分系数和种群规模对算法的影响进行了分析。实验证明取得了明显的优化效果，提高了神经网络的自适应能力和泛化能力，具有全局快速收敛的性能。

Optimization of neural networks with pseudo-parallelism genetic algorithm

A new approach is proposed,which combines pseudo-parallelism evolution technique based on sub-population competition with parent mutation mechanism,for automatic topology optimization of multi-layer feedforword neural networks.It's coding mode is a mix one concerning network layers and it allows that two networks with different number of units can be crossed to a new valid "child" network.The calculation result of computer simulations,an example of N-Parity,shows this hybrid algorithm,PPGA-MBP,is able to get the real-time information of population diversity during the process of evolution and has some improvements in both global conver-ging velocity and searching precision.