Cross-domain Contrastive Learning for Unsupervised Domain Adaptation
About
Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a fully-labeled source domain to a different unlabeled target domain. Most existing UDA methods learn domain-invariant feature representations by minimizing feature distances across domains. In this work, we build upon contrastive self-supervised learning to align features so as to reduce the domain discrepancy between training and testing sets. Exploring the same set of categories shared by both domains, we introduce a simple yet effective framework CDCL, for domain alignment. In particular, given an anchor image from one domain, we minimize its distances to cross-domain samples from the same class relative to those from different categories. Since target labels are unavailable, we use a clustering-based approach with carefully initialized centers to produce pseudo labels. In addition, we demonstrate that CDCL is a general framework and can be adapted to the data-free setting, where the source data are unavailable during training, with minimal modification. We conduct experiments on two widely used domain adaptation benchmarks, i.e., Office-31 and VisDA-2017, for image classification tasks, and demonstrate that CDCL achieves state-of-the-art performance on both datasets.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Classification | C16 | Balanced Accuracy66.6 | 31 | |
| Classification | CAMELYON17 Center 3 | CL Score77.4 | 18 | |
| Classification | CAMELYON17 Center 0 | Classification Accuracy72.5 | 18 | |
| Classification | CAMELYON17 Center 1 | CL59.3 | 18 | |
| Classification | CAMELYON17 Center 2 | CL72.8 | 18 | |
| Classification | CAMELYON17 Center 4 | CL Score55.4 | 18 | |
| Localization | C16 Target Domain from GlaS → C16 | PxAP37.9 | 18 | |
| Localization | CAMELYON17 Center 0 | PxAP27.2 | 18 | |
| Localization | CAMELYON17 Center 3 | PxAP37.6 | 18 | |
| Classification | Average C16, C17-0, C17-1, C17-2, C17-3, C17-4 | Accuracy (CL Average)66.9 | 18 |