BriefGPT.xyz
Nov, 2017
MinimalRNN: 更易于解释和训练的循环神经网络
MinimalRNN: Toward More Interpretable and Trainable Recurrent Neural Networks
HTML
PDF
Minmin Chen
TL;DR
通过采用极简的更新规则，MinimalRNN在学习和测试上不仅高效，而且更易解释和训练。最后，通过考察输入输出的Jacobi矩阵，发现MinimalRNN能够捕捉比现有RNN结构更长程的依赖关系。
Abstract
We introduce
minimalrnn
, a new
recurrent neural network
architecture that achieves comparable performance as the popular gated RNNs with a simplified structure. It employs minimal updates within RNN, which not on
→