Recent advances in adversarial training for adversarial robustness

T Bai, J Luo, J Zhao, B Wen, Q Wang - arXiv preprint arXiv:2102.01356, 2021 - arxiv.org
Adversarial training is one of the most effective approaches defending against adversarial
examples for deep learning models. Unlike other defense strategies, adversarial training …

Trustworthy AI: From principles to practices

B Li, P Qi, B Liu, S Di, J Liu, J Pei, J Yi… - ACM Computing Surveys, 2023 - dl.acm.org
The rapid development of Artificial Intelligence (AI) technology has enabled the deployment
of various systems based on it. However, many current AI systems are found vulnerable to …

Cross-entropy loss functions: Theoretical analysis and applications

A Mao, M Mohri, Y Zhong - International conference on …, 2023 - proceedings.mlr.press
Cross-entropy is a widely used loss function in applications. It coincides with the logistic loss
applied to the outputs of a neural network, when the softmax is used. But, what guarantees …

LAS-AT: adversarial training with learnable attack strategy

X Jia, Y Zhang, B Wu, K Ma… - Proceedings of the …, 2022 - openaccess.thecvf.com
Adversarial training (AT) is always formulated as a minimax problem, of which the
performance depends on the inner optimization that involves the generation of adversarial …

Reflection backdoor: A natural backdoor attack on deep neural networks

Y Liu, X Ma, J Bailey, F Lu - Computer Vision–ECCV 2020: 16th European …, 2020 - Springer
Recent studies have shown that DNNs can be compromised by backdoor attacks crafted at
training time. A backdoor attack installs a backdoor into the victim model by injecting a …

Neural attention distillation: Erasing backdoor triggers from deep neural networks

Y Li, X Lyu, N Koren, L Lyu, B Li, X Ma - arXiv preprint arXiv:2101.05930, 2021 - arxiv.org
Deep neural networks (DNNs) are known vulnerable to backdoor attacks, a training time
attack that injects a trigger pattern into a small proportion of training data so as to control the …

Privacy and robustness in federated learning: Attacks and defenses

L Lyu, H Yu, X Ma, C Chen, L Sun… - IEEE transactions on …, 2022 - ieeexplore.ieee.org
As data are increasingly being stored in different silos and societies becoming more aware
of data privacy issues, the traditional centralized training of artificial intelligence (AI) models …

Adversarial weight perturbation helps robust generalization

D Wu, ST Xia, Y Wang - Advances in neural information …, 2020 - proceedings.neurips.cc
The study on improving the robustness of deep neural networks against adversarial
examples grows rapidly in recent years. Among them, adversarial training is the most …

Improving adversarial robustness requires revisiting misclassified examples

Y Wang, D Zou, J Yi, J Bailey, X Ma… - … conference on learning …, 2019 - openreview.net
Deep neural networks (DNNs) are vulnerable to adversarial examples crafted by
imperceptible perturbations. A range of defense techniques have been proposed to improve …

Understanding and improving fast adversarial training

M Andriushchenko… - Advances in Neural …, 2020 - proceedings.neurips.cc
A recent line of work focused on making adversarial training computationally efficient for
deep learning models. In particular, Wong et al.(2020) showed that $\ell_\infty $-adversarial …