Enhanced Label Propagation through Affinity Matrix Fusion for Source-Free Domain Adaptation

Li Guo, Yuxuan Xia, Shengjie Wang

Primary Area: transfer learning, meta learning, and lifelong learning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: transfer learning, source-free domain adaptation, label propagation

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Abstract: Source-free domain adaptation (SFDA) has gained significant attention as a method to transfer knowledge from a pre-trained model on source domains toward target domains without accessing the source data. Recent research in SFDA has predominately adopted a self-training paradigm, focusing on utilizing local consistency constraints to refine pseudo-labels during self-training. These constraints encourage similar predictions among samples residing in local neighborhoods. Despite their effectiveness, the importance of global consistency is often overlooked. Moreover, such self-training-based adaptation processes suffer from the "confirmation bias": models use self-generated sub-optimal pseudo-labels to guide their subsequent training, resulting in a loop of self-reinforcing errors. In this study, we address the global consistency limitation by employing a label propagation method that seamlessly enforces both local and global consistency, leading to more coherent label predictions within the target domain. To mitigate the "confirmation bias", we propose utilizing an affinity matrix derived from current and historical models during the label propagation process. This approach takes advantage of different snapshots of the model to obtain a more accurate representation of the underlying graph structure, significantly enhancing the efficacy of label propagation and resulting in more refined pseudo-labels. Extensive experiments prove the superiority of our approach over the existing methods by a large margin. Our findings not only highlight the significance of incorporating global consistency within the SFDA framework but also offer a novel approach to mitigate the confirmation bias that arises from the use of noisy pseudo-labels in the self-training paradigm.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 2586