What's in a Loss Function for Image Classification?

Simon Kornblith
Simon Kornblith
Cited by: 0|Bibtex|Views9
Other Links: arxiv.org

Abstract:

It is common to use the softmax cross-entropy loss to train neural networks on classification datasets where a single class label is assigned to each example. However, it has been shown that modifying softmax cross-entropy with label smoothing or regularizers such as dropout can lead to higher performance. This paper studies a variety o...More

Code:

Data:

Your rating :
0

 

Tags
Comments