Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Learning Multimodal Data Augmentation in Feature Space


Dec 29, 2022
Zichang Liu, Zhiqiang Tang, Xingjian Shi, Aston Zhang, Mu Li, Anshumali Shrivastava, Andrew Gordon Wilson

Add code


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

What do Vision Transformers Learn? A Visual Exploration


Dec 13, 2022
Amin Ghiasi, Hamid Kazemi, Eitan Borgnia, Steven Reich, Manli Shu, Micah Goldblum, Andrew Gordon Wilson, Tom Goldstein

Add code


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Chroma-VAE: Mitigating Shortcut Learning with Generative Classifiers


Nov 28, 2022
Wanqian Yang, Polina Kirichenko, Micah Goldblum, Andrew Gordon Wilson

Add code

* Presented at the 36th Conference on Neural Information Processing Systems (NeurIPS 2022) 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

PAC-Bayes Compression Bounds So Tight That They Can Explain Generalization


Nov 24, 2022
Sanae Lotfi, Marc Finzi, Sanyam Kapoor, Andres Potapczynski, Micah Goldblum, Andrew Gordon Wilson

Add code

* NeurIPS 2022. Code is available at https://github.com/activatedgeek/tight-pac-bayes 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Bayesian Optimization with Conformal Coverage Guarantees


Oct 25, 2022
Samuel Stanton, Wesley Maddox, Andrew Gordon Wilson

Add code

* For code, see https://www.github.com/samuelstanton/conformal-bayesopt.git 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

K-SAM: Sharpness-Aware Minimization at the Speed of SGD


Oct 23, 2022
Renkun Ni, Ping-yeh Chiang, Jonas Geiping, Micah Goldblum, Andrew Gordon Wilson, Tom Goldstein

Add code

* 13 pages, 2 figures 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

On Feature Learning in the Presence of Spurious Correlations


Oct 20, 2022
Pavel Izmailov, Polina Kirichenko, Nate Gruver, Andrew Gordon Wilson

Add code

* NeurIPS 2022. Code available at https://github.com/izmailovpavel/spurious_feature_learning 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

How Much Data Are Augmentations Worth? An Investigation into Scaling Laws, Invariance, and Implicit Regularization


Oct 12, 2022
Jonas Geiping, Micah Goldblum, Gowthami Somepalli, Ravid Shwartz-Ziv, Tom Goldstein, Andrew Gordon Wilson

Add code

* 27 pages, 20 figures. Code at https://github.com/JonasGeiping/dataaugs 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

The Lie Derivative for Measuring Learned Equivariance


Oct 06, 2022
Nate Gruver, Marc Finzi, Micah Goldblum, Andrew Gordon Wilson

Add code


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Low-Precision Arithmetic for Fast Gaussian Processes


Jul 14, 2022
Wesley J. Maddox, Andres Potapczynski, Andrew Gordon Wilson

Add code

* UAI 2022. Code available at https://github.com/AndPotap/halfpres_gps 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email
1
2
3
4
5
6
7
>>