Search Results for author:
Found 2 papers, 0 papers with code
Broad Adversarial Training with Data Augmentation in the Output Space
Using B-AT instead of AT for supervised learning, we can increase the robustness by 0. 56\% for small seen attacks.
Noisy Adversarial Training
When combining a larger corpus of input data with our proposed method, we report an increase of the clean accuracy and for all observed attacks, compared to AT.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
