摘要
本文将动量项引入到训练Pi-sigma神经网络的异步批处理的梯度算法中,有效的改善了算法的收敛效率,并从理论上对该算法的收敛性进行研究,给出了误差函数的单调性定理及该算法的弱收敛和强收敛性定理.计算机仿真实验亦验证了带动量项的异步批处理梯度算法的有效性和理论分析的正确性.
An asynchronous gradient method with momentum for pi-sigma is considered. The learning rate is set to be a constant and the momentum factor an adaptive variable. Both the weak and strong convergence results are proved,as well as the convergence rates for the error function and for the weight. We also present numerical experiment to support our resuits.
出处
《应用数学》
CSCD
北大核心
2008年第1期207-212,共6页
Mathematica Applicata
基金
国家自然科学基金资助项目(10471017)
