In Neural Networks and deep neural networks what does label-dropout mean

Question

In Neural Networks and deep neural networks what does label-dropout mean

Michael

2022年2月6日 12:40

If you take the following sentence from an article on deep neural networks

to regularize the classifier layer by estimating the marginalized effect of label-dropout during training.

What does label-dropout mean

Topic dropout terminology deep-learning neural-network classification

Category Data Science

Yang Liu · Accepted Answer · 2022年2月6日 12:40

1

Yang Liu answered at 2022年2月6日 12:40

Perhaps it means groundtruth just keep one correct label while dropping out others. And this can cause some side-effects under extreme circumstances.

Valentin Calomme · Accepted Answer · 2017年10月15日 13:51

I ended up finding the paper you were referencing, maybe next time, add it to your post.

From what I can understand in this paper, label-dropout means that you are dropping the real labels and replace them with others.

The section of the paper you're referring to is explaining everything in great detail, so I'd try to read it carefully. In short, instead of using the ground-truth distribution (i.e. one 1 and a bunch of 0s) in the target layer, they are adding some uncertainty. That way, the model becomes less confident, and therefore is less prone to overfitting and becomes more adaptable.

In Neural Networks and deep neural networks what does label-dropout mean

About