ToAlign: Task-oriented Alignment for Unsupervised Domain Adaptation
arXiv:2106.10812
Abstract
Unsupervised domain adaptive classifcation intends to improve the classifcation performance on unlabeled target domain. To alleviate the adverse effect of domain shift, many approaches align the source and target domains in the feature space. However, a feature is usually taken as a whole for alignment without explicitly making domain alignment proactively serve the classifcation task, leading to sub-optimal solution. In this paper, we propose an effective Task-oriented Alignment (ToAlign) for unsupervised domain adaptation (UDA). We study what features should be aligned across domains and propose to make the domain alignment proactively serve classifcation by performing feature decomposition and alignment under the guidance of the prior knowledge induced from the classifcation task itself. Particularly, we explicitly decompose a feature in the source domain into a task-related/discriminative feature that should be aligned, and a task-irrelevant feature that should be avoided/ignored, based on the classifcation meta-knowledge. Extensive experimental results on various benchmarks (e.g., Offce-Home, Visda-2017, and DomainNet) under different domain adaptation settings demonstrate the effectiveness of ToAlign which helps achieve the state-of-the-art performance. The code is publicly available at https://github.com/microsoft/UDA
Accepted to NeurIPS 2021
References in corpus (6)
- VisDA: The Visual Domain Adaptation Challenge
- Domain Agnostic Learning with Disentangled Representations
- Learning Disentangled Semantic Representation for Domain Adaptation
- Feature Alignment and Restoration for Domain Generalization and Adaptation
- Your Classifier can Secretly Suffice Multi-Source Domain Adaptation
- Heuristic Domain Adaptation