A simple baseline for domain adaptation using rotation prediction
This work addresses domain adaptation for machine learning applications where labeled data is scarce, offering a simple and effective solution that outperforms existing methods in certain scenarios.
The paper tackles domain adaptation by proposing a self-supervised method using rotation prediction and self-distillation, achieving state-of-the-art results on the DomainNet dataset and showing robustness to bias in unlabeled data where other methods fail.
Recently, domain adaptation has become a hot research area with lots of applications. The goal is to adapt a model trained in one domain to another domain with scarce annotated data. We propose a simple yet effective method based on self-supervised learning that outperforms or is on par with most state-of-the-art algorithms, e.g. adversarial domain adaptation. Our method involves two phases: predicting random rotations (self-supervised) on the target domain along with correct labels for the source domain (supervised), and then using self-distillation on the target domain. Our simple method achieves state-of-the-art results on semi-supervised domain adaptation on DomainNet dataset. Further, we observe that the unlabeled target datasets of popular domain adaptation benchmarks do not contain any categories apart from testing categories. We believe this introduces a bias that does not exist in many real applications. We show that removing this bias from the unlabeled data results in a large drop in performance of state-of-the-art methods, while our simple method is relatively robust.