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

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Anish Athalye

Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks


Apr 08, 2021
Curtis G. Northcutt, Anish Athalye, Jonas Mueller

* ICLR 2021 RobustML and Weakly Supervised Learning Workshops; NeurIPS 2020 Workshop on Dataset Curation and Security 

  Access Paper or Ask Questions

On Evaluating Adversarial Robustness


Feb 20, 2019
Nicholas Carlini, Anish Athalye, Nicolas Papernot, Wieland Brendel, Jonas Rauber, Dimitris Tsipras, Ian Goodfellow, Aleksander Madry, Alexey Kurakin

* Living document; source available at https://github.com/evaluating-adversarial-robustness/adv-eval-paper/ 

  Access Paper or Ask Questions

Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples


Jul 31, 2018
Anish Athalye, Nicholas Carlini, David Wagner

* ICML 2018. Source code at https://github.com/anishathalye/obfuscated-gradients 

  Access Paper or Ask Questions

Evaluating and Understanding the Robustness of Adversarial Logit Pairing


Jul 26, 2018
Logan Engstrom, Andrew Ilyas, Anish Athalye

* Source code at https://github.com/labsix/adversarial-logit-pairing-analysis 

  Access Paper or Ask Questions

Black-box Adversarial Attacks with Limited Queries and Information


Jul 11, 2018
Andrew Ilyas, Logan Engstrom, Anish Athalye, Jessy Lin

* ICML 2018. This supercedes the previous paper "Query-efficient Black-box adversarial examples." 

  Access Paper or Ask Questions

Synthesizing Robust Adversarial Examples


Jun 07, 2018
Anish Athalye, Logan Engstrom, Andrew Ilyas, Kevin Kwok

* ICML 2018 

  Access Paper or Ask Questions

On the Robustness of the CVPR 2018 White-Box Adversarial Example Defenses


Apr 10, 2018
Anish Athalye, Nicholas Carlini


  Access Paper or Ask Questions

Query-Efficient Black-box Adversarial Examples (superceded)


Apr 06, 2018
Andrew Ilyas, Logan Engstrom, Anish Athalye, Jessy Lin

* Superceded by "Black-Box Adversarial Attacks with Limited Queries and Information." 

  Access Paper or Ask Questions