RCNN

Model

Recurrent Convolutional Neural Networks for Text Classification阅读笔记

Word Representation Learning

使用的双向RNN
cl(wi)c_l(w_i)表示词wiw_i左边上下文, cr(wi)c_r(w_i)表示右边上下文, 计算方式如下:

Recurrent Convolutional Neural Networks for Text Classification阅读笔记

e(wi1)e(w_{i-1})代表第i-1个词的word embedding.
每个词的最终表示:

Recurrent Convolutional Neural Networks for Text Classification阅读笔记

再接一个全连接作为输出:

Recurrent Convolutional Neural Networks for Text Classification阅读笔记

Text Representation Learning

对上一层的输出做max-pooling

Recurrent Convolutional Neural Networks for Text Classification阅读笔记
Recurrent Convolutional Neural Networks for Text Classification阅读笔记
Recurrent Convolutional Neural Networks for Text Classification阅读笔记

实验

参数设置

optimizer: SGD
learning rate: 0.01
embedding: 50
hidden layer(y(3)y^{(3)}的维度): 100

相关文章:

  • 2021-11-16
  • 2022-01-23
  • 2022-01-09
  • 2021-06-17
  • 2021-04-15
  • 2021-12-01
  • 2021-06-13
猜你喜欢
  • 2021-07-19
  • 2021-05-24
  • 2021-04-30
  • 2021-11-27
  • 2021-07-05
  • 2021-11-24
相关资源
相似解决方案