Convergence of Asynchronous Batch Gradient Method with Momentum for Pi-sigma Networks
Chao Zhang · Mathematica Applicata · 2008
An asynchronous gradient method with momentum for pisigma 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 results.