摘要
从多层感知器原理分析出发，提出一种自适应学习速率因子方法，用于对多层感知器中ＢＰ算法的改进，并将改进算法用于ＸＯＲ问题的学习及某分类器实例样本的学习。仿真结果表明，改进的ＢＰ算法可显著加速网络训练速度，学习过程具有较好的收敛性和较强的鲁棒性。
Based on the mechanism of multi-layer perception (MLP), a method with adaptive learning rate factors for the improvement of BP algorithm is presented. The improved algorithm is applied to the learning process of the XOR question and some classifier. The simulation results show the presented quick training algorithm can speed up the learning process of MLP, and improve the learning properties on convergence and robust performance.
出处
《控制与决策》
EI
CSCD
北大核心
2000年第1期125-127,共3页
Control and Decision
基金
山东省教委科研基金项目!(J98F04)
关键词
多层感知器
BP算法
网络训练法
神经网络
multi-layer perception, BP algorithm, adaptive learning rate factor
