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

Add to Chrome

Add to Firefox

Picture for Alexey Kurakin

Alexey Kurakin

Publishing Efficient On-device Models Increases Adversarial Vulnerability

Dec 28, 2022
Sanghyun Hong, Nicholas Carlini, Alexey Kurakin

Figure 1 for Publishing Efficient On-device Models Increases Adversarial Vulnerability

Figure 2 for Publishing Efficient On-device Models Increases Adversarial Vulnerability

Figure 3 for Publishing Efficient On-device Models Increases Adversarial Vulnerability

Figure 4 for Publishing Efficient On-device Models Increases Adversarial Vulnerability

* Accepted to IEEE SaTML 2023

Access Paper or Ask Questions

Toward Training at ImageNet Scale with Differential Privacy

Feb 09, 2022
Alexey Kurakin, Shuang Song, Steve Chien, Roxana Geambasu, Andreas Terzis, Abhradeep Thakurta

Figure 1 for Toward Training at ImageNet Scale with Differential Privacy

Figure 2 for Toward Training at ImageNet Scale with Differential Privacy

Figure 3 for Toward Training at ImageNet Scale with Differential Privacy

Figure 4 for Toward Training at ImageNet Scale with Differential Privacy

* 25 pages, 7 figures. Code available at https://github.com/google-research/dp-imagenet

Access Paper or Ask Questions

Handcrafted Backdoors in Deep Neural Networks

Jun 08, 2021
Sanghyun Hong, Nicholas Carlini, Alexey Kurakin

Figure 1 for Handcrafted Backdoors in Deep Neural Networks

Figure 2 for Handcrafted Backdoors in Deep Neural Networks

Figure 3 for Handcrafted Backdoors in Deep Neural Networks

Figure 4 for Handcrafted Backdoors in Deep Neural Networks

* 16 pages, 13 figures, 11 tables

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