BriefGPT.xyz
Dec, 2015
卷积神经网络正则化的最大池化减少丢失方法
Max-Pooling Dropout for Regularization of Convolutional Neural Networks
HTML
PDF
Haibing Wu, Xiaodong Gu
TL;DR
这篇论文研究了深度学习中dropout技术在pooling layer中的应用,提出了一种基于概率加权的pooling方法,称为probabilistic weighted pooling,并通过实验证明了它的优越性。
Abstract
Recently,
dropout
has seen increasing use in
deep learning
. For deep convolutional
neural networks
,
→