Timezone: »

 
Poster
Adversarial Training with Complementary Labels: On the Benefit of Gradually Informative Attacks
Jianan Zhou · Jianing Zhu · Jingfeng ZHANG · Tongliang Liu · Gang Niu · Bo Han · Masashi Sugiyama

@

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.

Keywords: adversarial training · weakly supervised learning · complementary label
[ Paper]  [ Poster]  [ OpenReview] 

Author Information

Jianan Zhou (Nanyang Technological University)
Jianing Zhu (HKBU)
Jingfeng ZHANG (RIKEN AIP)
Tongliang Liu (The University of Sydney)
Gang Niu (RIKEN)
Gang Niu

Gang Niu is currently an indefinite-term senior research scientist at RIKEN Center for Advanced Intelligence Project.

Bo Han (HKBU / RIKEN)
Masashi Sugiyama (RIKEN / University of Tokyo)

More from the Same Authors