改进交叉操作的遗传算法在神经网络优化中的应用

提出了一种对交叉操作进行改进的遗传算法优化神经网络的方法,它保留了父代中的优良模式,增加了找到最优解的概率,从而可以加快算法的收敛速度,缩短了寻找最优解的时间。实验证明,使用这种改进的交叉操作后的算法可以提高神经网络获得最优解的速度。

Genetic Algorithm Based on Modified Crossover Operator for Neural Network Optimization

This paper presents an improved genetic algorithm with a modified crossover operator for neural network optimization.It can maintain the high quality mode from the father generation and increase the possibility of finding the best answer,and using the modified crossover operator can speed up the convergence rate,thus shorting time to find the best answer.The experimental result indicates that this method can speed up the rate of finding the best answer for neural network.