Mikolov T, Sutskever I, Chen K, et al. Distributed representations of words and phrases and their compositionality[C]// International Conference on Neural Information Processing Systems. Curran Associates Inc. 2013:3111-3119.
context | word | target |
---|---|---|
orange | juice | 1 |
orange | king | 0 |
orange | book | 0 |
orange | the | 0 |
orange | of | 0 |
Skip-grams 中softmax函数定义:
\[P(target|content)=\frac{e^{\theta_{t}^{T}e_{c}}}{\sum^{10000}_{j=1}{e^{\theta^{T}_{j}e_{c}}}}\]
本节算法定义 输入Context为c,Word为t,定义输出Target为y
context | word | target |
---|---|---|
c | t | y |
\(x_1\) | \(x_2\) | y |
orange | juice | 1 |
orange | king | 0 |
orange | book | 0 |
orange | the | 0 |
orange | of | 0 |
[DeeplearningAI笔记]序列模型2.7负采样Negative sampling
原文:https://www.cnblogs.com/cloud-ken/p/9556278.html