Skip to content

We investigated corruption robustness across different architectures including Convolutional Neural Networks, Vision Transformers, and the MLP-Mixer.

Notifications You must be signed in to change notification settings

katelyn98/CorruptionRobustness

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 

Repository files navigation

CorruptionRobustness

Open In Colab

Paper

This work presented at the Uncertainty & Robustness in Deep Learning workshop at ICML 2021. You can view our current arXiv paper here. Note: some of the normalization techniques we used we later found out were not appropriate for certain models. However, originally, normalziation for all models was the same values (imagenet normalization). This is a workshop paper and is preliminary work done by all undergraduates. We appreciate the kind feedback.

Summary

We explored corruption robustness across different Convolutional Neural Networks, Vision Transformer architectures, and the MLP-Mixer.

Corruption Robustness

Open In Colab

Shape Bias

Open In Colab

Analysis

Coming soon - view appendix in the meantime!

About

We investigated corruption robustness across different architectures including Convolutional Neural Networks, Vision Transformers, and the MLP-Mixer.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published