BriefGPT.xyz
May, 2020
关于均场极限下两层ReLU网络梯度下降训练的收敛性
On the Convergence of Gradient Descent Training for Two-layer ReLU-networks in the Mean Field Regime
HTML
PDF
Stephan Wojtowytsch
TL;DR
在均场模型下，用梯度下降训练双层ReLU网络时，我们描述了最小贝叶斯风险收敛的必要和充分条件。该条件不依赖于参数的初始化，只涉及神经网络实现的弱收敛，而不是其参数分布。
Abstract
We describe a necessary and sufficient condition for the
convergence
to minimum
bayes risk
when training two-layer
relu-networks
by
→