文献类型：会议论文

英文题名：Variable-step BP training algorithm with an adaptive momentum term

作者：Yang, Huizhong[1]; Li, Danjing[1]; Tao, Zhenlin[1]; Zhang, Suzhen[1]

机构：[1]E China Univ Sci & Technol, Res Inst Automat, Shanghai 200237, Peoples R China

会议论文集：4th World Congress on Intelligent Control and Automation

会议日期：JUN 10-14, 2002

会议地点：SHANGHAI, PEOPLES R CHINA

语种：英文

外文关键词：variable-step BP training algorithm; adaptive momentum term; temperature ascending strategy; temperature descending strategy

摘要：This paper proposes a new variable-step BP training algorithm which can overcome some disadvantages in the traditional BP training algorithm such as slow convergent speed and tendency of getting in local minima. In the new method, a nonlinear property term is introduced and its intensity coefficient is constructed as an adaptive nonlinear function, which processes temperature ascending and temperature descending strategy. The simulation results show the efficiency of the new method.