This work conducts the first analysis on the robustness against adversarial attacks on self-supervised Vision Transformers trained using DINO. First, we evaluate whether features learned through self-supervision are more robust to adversarial attacks than those emerging from supervised learning. Then, we present properties arising for attacks in the latent space. Finally, we evaluate whether three well-known defense strategies can increase adversarial robustness in downstream tasks by only fine-tuning the classification head to provide robustness even in view of limited compute resources. These defense strategies are: Adversarial Training, Ensemble Adversarial Training and Ensemble of Specialized Networks.
@article{arxiv.2206.06761,
title = {Exploring Adversarial Attacks and Defenses in Vision Transformers trained with DINO},
author = {Javier Rando and Nasib Naimi and Thomas Baumann and Max Mathys},
journal= {arXiv preprint arXiv:2206.06761},
year = {2022}
}
Comments
ICML 2022 Workshop paper accepted at AdvML Frontiers