Skip to content

xidui/Multivariate-Gaussian-Dropout

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

43 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Multivariate Gaussian Dropout

Project for 10-707: Deep Learning @ Carnegie Mellon University.

Proposed and implemented a new dropout method that generates the mask by sampling from a multivariate Gaussian distribution, whose parameters are dynamically estimated from statistics calculated over a mini-batch of training data.

Conducted experiments on MNIST & CIFAR-10 datasets and showed that the proposed method can improve performance in terms of both convergence speed and generalization capability.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages