Gagandeep Singh

activity

20242026

most citedTowards Generalized Certified Robustness with Multi-Norm Training

1 citations · 1 across the 4 of their papers we have counts for

collaborators

Showing cs.LGShow all

2 papers · 1 filter

cs.LG2024★ 1 cited

Towards Generalized Certified Robustness with Multi-Norm Training

Enyi Jiang, David S. Cheung, Gagandeep Singh

Existing certified training methods can only train models to be robust against a certain perturbation type (e.g. $l_{\infty}$ or $l_{2}$ ). However, an $l_{\infty}$ certifiably robust mod…

cs.LG2024

RAMP: Boosting Adversarial Robustness Against Multiple $l_{p}$ Perturbations for Universal Robustness

Enyi Jiang, Gagandeep Singh

Most existing works focus on improving robustness against adversarial attacks bounded by a single $l_{p}$ norm using adversarial training (AT). However, these AT models' multiple-nor…

Towards Generalized Certified Robustness with Multi-Norm Training

RAMP: Boosting Adversarial Robustness Against Multiple lp​ Perturbations for Universal Robustness

RAMP: Boosting Adversarial Robustness Against Multiple $l_{p}$ Perturbations for Universal Robustness