Search Results for author:
Found 4 papers, 0 papers with code
AdvFunMatch: When Consistent Teaching Meets Adversarial Robustness
To tackle this problem, we propose a simple but effective strategy called Adversarial Function Matching (AdvFunMatch), which aims to match distributions for all data points within the $\ell_p$-norm ball of the training data, in accordance with consistent teaching.
Lower Difficulty and Better Robustness: A Bregman Divergence Perspective for Adversarial Training
In addition, we discuss the function of entropy in TRADES, and we find that models with high entropy can be better robustness learners.
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.
Understanding the robustness-accuracy tradeoff by rethinking robust fairness
Then, we explored the effect of another classic smoothing regularizer, namely, the maximum entropy (ME), and we have found ME can also help reduce both inter-class similarity and intra-class variance.
