Understanding robust overfitting of adversarial training and beyond
Robust overfitting widely exists in adversarial training of deep networks. The exact
underlying reasons for this are still not completely understood. Here, we explore the causes …
underlying reasons for this are still not completely understood. Here, we explore the causes …
Revisiting adversarial robustness distillation from the perspective of robust fairness
Abstract Adversarial Robustness Distillation (ARD) aims to transfer the robustness of large
teacher models to small student models, facilitating the attainment of robust performance on …
teacher models to small student models, facilitating the attainment of robust performance on …
Scaling adversarial training to large perturbation bounds
Abstract The vulnerability of Deep Neural Networks to Adversarial Attacks has fuelled
research towards building robust models. While most Adversarial Training algorithms aim at …
research towards building robust models. While most Adversarial Training algorithms aim at …
Towards achieving adversarial robustness beyond perceptual limits
The vulnerability of Deep Neural Networks to Adversarial Attacks has fuelled research
towards building robust models. While most Adversarial Training algorithms aim towards …
towards building robust models. While most Adversarial Training algorithms aim towards …
Revisiting Adversarial Training under Long-Tailed Distributions
Deep neural networks are vulnerable to adversarial attacks leading to erroneous outputs.
Adversarial training has been recognized as one of the most effective methods to counter …
Adversarial training has been recognized as one of the most effective methods to counter …
Memorization weights for instance reweighting in adversarial training
Adversarial training is an effective way to defend deep neural networks (DNN) against
adversarial examples. However, there are atypical samples that are rare and hard to learn …
adversarial examples. However, there are atypical samples that are rare and hard to learn …
Towards transferable unrestricted adversarial examples with minimum changes
Transfer-based adversarial example is one of the most important classes of black-box
attacks. However, there is a trade-off between transferability and imperceptibility of the …
attacks. However, there is a trade-off between transferability and imperceptibility of the …
One-vs-the-rest loss to focus on important samples in adversarial training
This paper proposes a new loss function for adversarial training. Since adversarial training
has difficulties, eg, necessity of high model capacity, focusing on important data points by …
has difficulties, eg, necessity of high model capacity, focusing on important data points by …
On the impact of hard adversarial instances on overfitting in adversarial training
Adversarial training is a popular method to robustify models against adversarial attacks.
However, it exhibits much more severe overfitting than training on clean inputs. In this work …
However, it exhibits much more severe overfitting than training on clean inputs. In this work …
Alleviating robust overfitting of adversarial training with consistency regularization
Adversarial training (AT) has proven to be one of the most effective ways to defend Deep
Neural Networks (DNNs) against adversarial attacks. However, the phenomenon of robust …
Neural Networks (DNNs) against adversarial attacks. However, the phenomenon of robust …