fra31

auto-attack. Code relative to "Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks"

sparse-rs. Sparse-RS: a versatile framework for query-efficient sparse black-box adversarial attacks

sparse-imperceivable-attacks. Sparse and Imperceivable Adversarial Attacks (accepted to ICCV 2019).

fab-attack. Code for FAB-attack

evaluating-adaptive-test-time-defenses. Python

robust-finetuning. Code relative to "Adversarial robustness against multiple and single $l_p$-threat models via quick fine-tuning of robust classifiers"

rlhf-trojan-competition-submission. Python

perceptual-metrics. Python

mmr-universal. Code for training with MMR-Universal regularizer.

advertorch. A Toolbox for Adversarial Robustness Research