摘要
Adversarial examples causing evasive predictions are widely used to evaluate and improve the robustness of machine learning models. However, current studies focus on supervised learning tasks, relying on the ground-truth data label, a targeted objective, or supervision from a trained classifier. In this paper, we propose a framework of generating adversarial examples for unsupervised models and demonstrate novel applications to data augmentation. Our framework exploits a mutual information neural estimator as an information-theoretic similarity measure to generate adversarial examples without supervision. We propose a new MinMax algorithm with provable convergence guarantees for efficient generation of unsupervised adversarial examples. When using unsupervised adversarial examples as a simple plug-in data augmentation tool for model retraining, significant improvements are consistently observed across different unsupervised tasks and datasets, including data reconstruction, representation learning, and contrastive learning.
原文 | American English |
---|---|
出版狀態 | Published - 5月 2021 |
事件 | The International Conference on Learning Representations (ICLR) 2021 - 持續時間: 4 5月 2021 → 7 5月 2021 |
Workshop
Workshop | The International Conference on Learning Representations (ICLR) 2021 |
---|---|
期間 | 4/05/21 → 7/05/21 |