Rethinking Image Mixture for Unsupervised Visual Representation Learning

Cited by: 0|Bibtex|Views121
Other Links: arxiv.org

Abstract:

In supervised learning, smoothing label/prediction distribution in neural network training has been proven useful in preventing the model from being over-confident, and is crucial for learning more robust visual representations. This observation motivates us to explore the way to make predictions flattened in unsupervised learning. Cons...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments