BriefGPT.xyz
Aug, 2015
基于嵌入式神经网络的正则化策略比较研究
A Comparative Study on Regularization Strategies for Embedding-based Neural Networks
HTML
PDF
Hao Peng, Lili Mou, Ge Li, Yunchuan Chen, Yangyang Lu...
TL;DR
本文旨在比较不同的正则化策略，以解决嵌入式神经网络在NLP中严重过拟合的现象。研究着重于超参数调整和组合不同的正则化策略，结果提供了神经NLP模型的超参数调整图片。
Abstract
This paper aims to compare different
regularization strategies
to address a common phenomenon, severe overfitting, in
embedding-based neural networks
for
→