We seek to learn a representation that captures the shared information between clean samples and their corresponding adversarial samples.

2022/11/02 · Abstract. We consider the problem of improving the adversarial robustness of neural networks while retaining natural accuracy.

We consider the problem of improving the adversarial robustness of neural networks while retaining natural accuracy. Motivated by the multi-view information ...

We consider the problem of improving the adversarial robustness of neural networks while retaining natural accuracy. Motivated by the multi-view information ...

2023/06/27 · In this paper, we propose a novel training framework called Robust Proxy Learning. In the proposed method, the model explicitly learns robust feature ...

We propose to enhance the adversarial robustness by maximizing the natural MI and minimizing the adversarial MI during the training process.

This study investigates the impact of such closely-coupled classes on adversarial attacks and develops a self-paced reweighting strategy in adversarial training ...

In this paper, we propose a novel ensemble approach to improve the robustness of classifiers against evasion attacks by using diversified feature selection and ...

The experimental results demonstrate that augmenting adversarial training with our proposed components can further improve the robustness of the network, ...

2023/03/20 · We propose a novel defense algorithm called Between-Class Adversarial Training (BCAT) that combines Between-Class learning (BC-learning) with standard AT.