摘要
提出用实数编码的遗传算法来优化多层神经网络的权值,并且将遗传算法与BP算法结合,能有效地避免BP算法陷入局部极小和遗传算法过早收敛,实验结果令人满意.
In this paper genetic algorithm coding in real number is engaged to optimize the weights of multi layer neural network. Furthermore, by combining this real coded GA with BP algorithm, the entrapment in local optical optimum of BP and the premature of GA can be prevented efficiently. Some satisfactory results are obtained.
出处
《武汉大学学报(自然科学版)》
CSCD
1998年第3期289-291,共3页
Journal of Wuhan University(Natural Science Edition)
基金
国家自然科学基金
教育部博士点基金
关键词
神经网络
遗传算法
BP算法
编码
解码
neural network, genetic algorithm, backpropagation algorithm, XOR problem, encode/decode problem