Corpus ID: 233004251

Curriculum Graph Co-Teaching for Multi-Target Domain Adaptation

@article{Roy2021CurriculumGC,
  title={Curriculum Graph Co-Teaching for Multi-Target Domain Adaptation},
  author={Subhankar Roy and Evgeny Krivosheev and Zhun Zhong and N. Sebe and E. Ricci},
  journal={ArXiv},
  year={2021},
  volume={abs/2104.00808}
}
In this paper we address multi-target domain adaptation (MTDA), where given one labeled source dataset and multiple unlabeled target datasets that differ in data distributions, the task is to learn a robust predictor for all the target domains. We identify two key aspects that can help to alleviate multiple domain-shifts in the MTDA: feature aggregation and curriculum learning. To this end, we propose Curriculum Graph Co-Teaching (CGCT) that uses a dual classifier head, with one of them being a… Expand

References

SHOWING 1-10 OF 61 REFERENCES
Unsupervised Multi-Target Domain Adaptation Through Knowledge Distillation
Progressive Feature Alignment for Unsupervised Domain Adaptation
Heterogeneous Graph Attention Network for Unsupervised Multiple-Target Domain Adaptation.
Blending-Target Domain Adaptation by Adversarial Meta-Adaptation Networks
Moment Matching for Multi-Source Domain Adaptation
Open Compound Domain Adaptation
  • Z. Liu, Zhongqi Miao, +4 authors Boqing Gong
  • Computer Science, Mathematics
  • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2020
GCAN: Graph Convolutional Adversarial Network for Unsupervised Domain Adaptation
  • Xinhong Ma, T. Zhang, C. Xu
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
...
1
2
3
4
5
...