English

Active Learning for Domain Adaptation: An Energy-Based Approach

Machine Learning 2022-03-10 v3 Computer Vision and Pattern Recognition

Abstract

Unsupervised domain adaptation has recently emerged as an effective paradigm for generalizing deep neural networks to new target domains. However, there is still enormous potential to be tapped to reach the fully supervised performance. In this paper, we present a novel active learning strategy to assist knowledge transfer in the target domain, dubbed active domain adaptation. We start from an observation that energy-based models exhibit \textit{free energy biases} when training (source) and test (target) data come from different distributions. Inspired by this inherent mechanism, we empirically reveal that a simple yet efficient energy-based sampling strategy sheds light on selecting the most valuable target samples than existing approaches requiring particular architectures or computation of the distances. Our algorithm, Energy-based Active Domain Adaptation (EADA), queries groups of target data that incorporate both domain characteristic and instance uncertainty into every selection round. Meanwhile, by aligning the free energy of target data compact around the source domain via a regularization term, domain gap can be implicitly diminished. Through extensive experiments, we show that EADA surpasses state-of-the-art methods on well-known challenging benchmarks with substantial improvements, making it a useful option in the open world. Code is available at https://github.com/BIT-DA/EADA.

Keywords

Cite

@article{arxiv.2112.01406,
  title  = {Active Learning for Domain Adaptation: An Energy-Based Approach},
  author = {Binhui Xie and Longhui Yuan and Shuang Li and Chi Harold Liu and Xinjing Cheng and Guoren Wang},
  journal= {arXiv preprint arXiv:2112.01406},
  year   = {2022}
}

Comments

Camera ready for AAAI 2022. Code is available at https://github.com/BIT-DA/EADA

R2 v1 2026-06-24T08:01:58.495Z