Published in Dec, 2018
This paper is about the adversarial robustness of deep neural nets.
Yan, Ziang*, Yiwen Guo*, and Changshui Zhang. "DeepDefense: Training Deep Neural Networks with Improved Robustness." NeurIPS 2018.
Published in Oct, 2019
We utilize adversarial perturbation as a proxy to improve input space margin of DNN
Yan, Ziang, Yiwen Guo, and Changshui Zhang. "Adversarial Margin Maximization Networks." PAMI 2019.
Published in Dec, 2019
This paper aims at reducing the query complexity of black-box adversarial attacks.
Yan, Ziang*, Yiwen Guo*, and Changshui Zhang. "Subspace Attack: Exploiting Promising Subspaces for Query-Efficient Black-box Attacks." NeurIPS 2019.
Published in Jan, 2021
We utilize policy learning techniques to reduce the query complexity of hard-label black-box adversarial attacks.
Yan, Ziang, Yiwen Guo, Jian Liang, and Changshui Zhang. "Policy-Driven Attack: Learning to Query for Hard-label Black-box Adversarial Examples." ICLR 2021.