Regularization

Label Smoothing

Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the likelihood of $\log{p}\left(y\mid{x}\right)$ directly can be harmful. Assume for a small constant $\epsilon$, the training set label $y$ is correct with probability $1-\epsilon$ and incorrect otherwise. Label Smoothing regularizes a model based on a softmax with $k$ output values by replacing the hard $0$ and $1$ classification targets with targets of $\frac{\epsilon}{k}$ and $1-\frac{k-1}{k}\epsilon$ respectively.

Source: Deep Learning, Goodfellow et al

Image Source: When Does Label Smoothing Help?

Papers


Paper Code Results Date Stars

Tasks


Task Papers Share
Large Language Model 23 2.79%
Computational Efficiency 23 2.79%
Language Modeling 21 2.55%
Language Modelling 21 2.55%
Decoder 19 2.31%
Question Answering 15 1.82%
Semantic Segmentation 14 1.70%
Image Generation 13 1.58%
Image Classification 12 1.46%

Components


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

Categories