摘要： 影响多层神经网络ＢＰ算法学习效率的因素不少．但ＢＰ算法中误差函数不能有效地表征样本学习精度是其中主要的因素之一．本文对ＢＰ算法中的误差函数进行了修正．计算机模拟结果表明这种修正有助于提高学习精度和学习效率．

关键词: 人工神经网络；BP算法；多层神经网络

Abstract: The cffcctivcncss of BP algorithm used in Multilayer Neural Network is influenced by many factors,one of which is that the error function can not express the error of the sample--learning cffcctivcly. The paper improves the original error function,and shows,by computer simulation,that the improved error function helps thesample-learning a lot.

中图分类号: 

• TP18