Rethinking the Role of Pre-Trained Networks in Source-Free Domain Adaptation
This work addresses domain adaptation for scenarios where only a pre-trained source model and unlabeled target data are available, offering an incremental improvement by leveraging pre-trained networks more effectively.
The paper tackles the problem of source-free domain adaptation (SFDA) by integrating pre-trained networks into the target adaptation process to prevent overfitting to source data and improve generalization, resulting in improved adaptation performance on 4 benchmark datasets.
Source-free domain adaptation (SFDA) aims to adapt a source model trained on a fully-labeled source domain to an unlabeled target domain. Large-data pre-trained networks are used to initialize source models during source training, and subsequently discarded. However, source training can cause the model to overfit to source data distribution and lose applicable target domain knowledge. We propose to integrate the pre-trained network into the target adaptation process as it has diversified features important for generalization and provides an alternate view of features and classification decisions different from the source model. We propose to distil useful target domain information through a co-learning strategy to improve target pseudolabel quality for finetuning the source model. Evaluation on 4 benchmark datasets show that our proposed strategy improves adaptation performance and can be successfully integrated with existing SFDA methods. Leveraging modern pre-trained networks that have stronger representation learning ability in the co-learning strategy further boosts performance.