Our Adversarial Training Paper has been Accepted by AISTATS 2021. We propose a learning-to-learn method to improve adversarial training on a variety of tasks.

Reference⌗

  • Learning to Defense by Learning to Attack 2021
    Zhehui Chen*, Haoming Jiang*, Yuyang Shi, Bo Dai, and Tuo Zhao (* Equal Contribution) [arXiv]
    The 24th International Conference on Artificial Intelligence and Statistics (AISTATS), 2021