Skip to content

The official repository for the experiments included in the paper titled "Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks" [ICML, 2023]

License

Notifications You must be signed in to change notification settings

nowazrabbani/pMoE_CNN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

68 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Code of the paper titled "Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks" [ICML, 2023]

by Mohammed Nowaz Rabbani Chowdhury, Shuai Zhang, Meng Wang, Sijia Liu and Pin-Yu Chen

Requirements

tensorflow, numpy, matplotlib

Description

For experiments on each dataset, there is one directory. For example, for all the experiments on the CelebA dataset, the directory is "CelebA Exp".

The data files need to be downloaded in the experiment directory using the hyperlinks provided at the file data_files_*.txt of the directory.

The main_*.py in each folder train the networks and plot the results.

About

The official repository for the experiments included in the paper titled "Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks" [ICML, 2023]

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages