今天发现了自己好多的知识盲区……待填坑——
CRF
参考博客：medium.com/ml2vec/over… 这是为数不多的我能看的进去的讲解了…… 虽然我还是没太看懂他的数学讲解部分……
记几个点吧：
1、Conditional Random Fields are a type of Discriminative classifier, and as such, they model the decision boundary between the different classes.
2、Hidden Markov Models are generative, and give output by modeling the joint probability distribution. On the other hand, Conditional Random Fields are discriminative, and model the conditional probability distribution.
3、CRFs don’t rely on the independence assumption (that the labels are independent of each other), and avoid label bias.
encoder-decoder模型的必要性
待填坑
Transformer
待填坑
loss函数的各种可选形式
待填坑
Skip connection的具体思维还需要了解
待填坑。
bilnear upsampling
待填坑。
Gumbel-Softmax
待填坑。
