RandomOut: Using a convolutional gradient norm to win The Filter Lottery on ShortScience.org

RandomOut: Using a convolutional gradient norm to win The Filter Lottery
Cohen, Joseph Paul and Lo, Henry Z. and Ding, Wei
arXiv e-Print archive - 2016 via Local Bibsonomy
Keywords: dblp

[link] Summary by Henry Z Lo 8 years ago

The proposed Randomout algorithm randomly restarts filter weights in a CNN when it has a low gradient.

This is a very short summary.

Your comment:

Write your summary here (You can use $\LaTeX$ and markdown syntax):

Anon Private