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

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Certified Defences Against Adversarial Patch Attacks on Semantic Segmentation


Sep 13, 2022
Maksym Yatsura, Kaspar Sakmann, N. Grace Hua, Matthias Hein, Jan Hendrik Metzen


   Access Paper or Ask Questions

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

Give Me Your Attention: Dot-Product Attention Considered Harmful for Adversarial Patch Robustness


Mar 25, 2022
Giulio Lovisotto, Nicole Finnie, Mauricio Munoz, Chaithanya Kumar Mummadi, Jan Hendrik Metzen

* to be published in IEEE/CVF Conference on Computer Vision and Pattern Recognition 2022, CVPR22 

   Access Paper or Ask Questions

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

Neural Architecture Search for Dense Prediction Tasks in Computer Vision


Feb 15, 2022
Thomas Elsken, Arber Zela, Jan Hendrik Metzen, Benedikt Staffler, Thomas Brox, Abhinav Valada, Frank Hutter


   Access Paper or Ask Questions

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

Meta-Learning the Search Distribution of Black-Box Random Search Based Adversarial Attacks


Nov 22, 2021
Maksym Yatsura, Jan Hendrik Metzen, Matthias Hein

* accepted at NeurIPS 2021; updated the numbers in Table 5 and added references; added acknowledgements 

   Access Paper or Ask Questions

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

Bag of Tricks for Neural Architecture Search


Jul 08, 2021
Thomas Elsken, Benedikt Staffler, Arber Zela, Jan Hendrik Metzen, Frank Hutter


   Access Paper or Ask Questions

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

Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformation


Jun 28, 2021
Chaithanya Kumar Mummadi, Robin Hutmacher, Kilian Rambach, Evgeny Levinkov, Thomas Brox, Jan Hendrik Metzen

* 16 pages, 5 figures, 7 tables 

   Access Paper or Ask Questions

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

Does enhanced shape bias improve neural network robustness to common corruptions?


Apr 20, 2021
Chaithanya Kumar Mummadi, Ranjitha Subramaniam, Robin Hutmacher, Julien Vitay, Volker Fischer, Jan Hendrik Metzen

* 20 pages, 9 figures, 12 tables, accepted at ICLR 2021 

   Access Paper or Ask Questions

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

Efficient Certified Defenses Against Patch Attacks on Image Classifiers


Feb 08, 2021
Jan Hendrik Metzen, Maksym Yatsura

* accepted at ICLR 2021 

   Access Paper or Ask Questions

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

Meta Adversarial Training


Jan 27, 2021
Jan Hendrik Metzen, Nicole Finnie, Robin Hutmacher


   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
>>