Regularization

Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is $p=0.5$). At test time, all units are present, but with weights scaled by $p$ (i.e. $w$ becomes $pw$).

The idea is to prevent co-adaptation, where the neural network becomes too reliant on particular connections, as this could be symptomatic of overfitting. Intuitively, dropout can be thought of as creating an implicit ensemble of neural networks.

Source: Dropout: A Simple Way to Prevent Neural Networks from Overfitting

Papers


Paper Code Results Date Stars

Tasks


Task Papers Share
Language Modelling 71 9.24%
Large Language Model 37 4.82%
Semantic Segmentation 23 2.99%
Retrieval 17 2.21%
Decision Making 17 2.21%
Prompt Engineering 14 1.82%
Question Answering 12 1.56%
Benchmarking 12 1.56%
Classification 12 1.56%

Components


Component Type
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories