Local Context-Aware Active Domain Adaptation
This work addresses the challenge of efficient model adaptation across domains with limited labeling budgets, which is incremental by building on existing ADA methods to better handle large domain gaps.
The paper tackles the problem of selecting informative target samples in Active Domain Adaptation (ADA) by proposing a local context-aware framework (LADA) that uses a novel criterion based on local inconsistency of model predictions and augments labeled data with confident neighbors. The method outperforms recent ADA approaches on various benchmarks, with concrete improvements in sample selection efficiency and adaptation performance.
Active Domain Adaptation (ADA) queries the labels of a small number of selected target samples to help adapting a model from a source domain to a target domain. The local context of queried data is important, especially when the domain gap is large. However, this has not been fully explored by existing ADA works. In this paper, we propose a Local context-aware ADA framework, named LADA, to address this issue. To select informative target samples, we devise a novel criterion based on the local inconsistency of model predictions. Since the labeling budget is usually small, fine-tuning model on only queried data can be inefficient. We progressively augment labeled target data with the confident neighbors in a class-balanced manner. Experiments validate that the proposed criterion chooses more informative target samples than existing active selection strategies. Furthermore, our full method clearly surpasses recent ADA arts on various benchmarks. Code is available at https://github.com/tsun/LADA.