Class-Aware Contrastive Semi-Supervised Learning
About
Pseudo-label-based semi-supervised learning (SSL) has achieved great success on raw data utilization. However, its training procedure suffers from confirmation bias due to the noise contained in self-generated artificial labels. Moreover, the model's judgment becomes noisier in real-world applications with extensive out-of-distribution data. To address this issue, we propose a general method named Class-aware Contrastive Semi-Supervised Learning (CCSSL), which is a drop-in helper to improve the pseudo-label quality and enhance the model's robustness in the real-world setting. Rather than treating real-world data as a union set, our method separately handles reliable in-distribution data with class-wise clustering for blending into downstream tasks and noisy out-of-distribution data with image-wise contrastive for better generalization. Furthermore, by applying target re-weighting, we successfully emphasize clean label learning and simultaneously reduce noisy label learning. Despite its simplicity, our proposed CCSSL has significant performance improvements over the state-of-the-art SSL methods on the standard datasets CIFAR100 and STL10. On the real-world dataset Semi-iNat 2021, we improve FixMatch by 9.80% and CoMatch by 3.18%. Code is available https://github.com/TencentYoutuResearch/Classification-SemiCLS.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Image Classification | CIFAR-100 (test) | Accuracy71.13 | 3518 | |
| Image Classification | CIFAR-10 (test) | Accuracy88.77 | 3381 | |
| Image Classification | CIFAR10 (test) | Accuracy95.54 | 585 | |
| Image Classification | SUN397 | -- | 425 | |
| Image Classification | DTD | Accuracy72.7 | 419 | |
| Image Classification | CIFAR100 (test) | Top-1 Accuracy80.68 | 377 | |
| Image Classification | STL-10 (test) | Accuracy82 | 357 | |
| Image Classification | CIFAR-100 | -- | 302 | |
| Image Classification | SVHN (test) | Accuracy88.6 | 199 | |
| Image Classification | Caltech-101 | -- | 146 |