Adversarial Training with Complementary Labels: On the Benefit of Gradually Informative Attacks
This work addresses adversarial training in practical scenarios with imperfect supervision, offering a novel approach for robust machine learning, though it is incremental in advancing existing adversarial training methods.
The paper tackles adversarial training with complementary labels, a challenging setting where labels specify what class a sample is not, and proposes a method using gradually informative attacks to address intractable optimization and low-quality adversarial examples, achieving improved performance on benchmark datasets.
Adversarial training (AT) with imperfect supervision is significant but receives limited attention. To push AT towards more practical scenarios, we explore a brand new yet challenging setting, i.e., AT with complementary labels (CLs), which specify a class that a data sample does not belong to. However, the direct combination of AT with existing methods for CLs results in consistent failure, but not on a simple baseline of two-stage training. In this paper, we further explore the phenomenon and identify the underlying challenges of AT with CLs as intractable adversarial optimization and low-quality adversarial examples. To address the above problems, we propose a new learning strategy using gradually informative attacks, which consists of two critical components: 1) Warm-up Attack (Warm-up) gently raises the adversarial perturbation budgets to ease the adversarial optimization with CLs; 2) Pseudo-Label Attack (PLA) incorporates the progressively informative model predictions into a corrected complementary loss. Extensive experiments are conducted to demonstrate the effectiveness of our method on a range of benchmarked datasets. The code is publicly available at: https://github.com/RoyalSkye/ATCL.