Regularization

Attention Dropout

Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the softmax in the attention equation. For example, for scaled-dot product attention, we would drop elements from the first term:

$$ {\text{Attention}}(Q, K, V) = \text{softmax}\left(\frac{QK^{T}}{\sqrt{d_k}}\right)V $$

Papers


Paper Code Results Date Stars

Tasks


Task Papers Share
Language Modelling 85 10.75%
Retrieval 64 8.09%
Question Answering 50 6.32%
Large Language Model 42 5.31%
Sentence 31 3.92%
Text Generation 23 2.91%
In-Context Learning 20 2.53%
Information Retrieval 16 2.02%
Text Classification 15 1.90%

Components


Component Type
Dropout
Regularization

Categories