Learning to Generalize: Meta-Learning for Domain Generalization

@inproceedings{Li2018LearningTG,
  title={Learning to Generalize: Meta-Learning for Domain Generalization},
  author={Da Li and Yongxin Yang and Yi-Zhe Song and Timothy M. Hospedales},
  booktitle={AAAI},
  year={2018}
}
Domain shift refers to the well known problem that a model trained in one source domain performs poorly when applied to a target domain with different statistics. Domain Generalization (DG) techniques attempt to alleviate this issue by producing models which by design generalize well to novel testing domains. We propose a novel meta-learning method for domain generalization. Rather than designing a specific model that is robust to domain shift as in most previous DG work, we propose a model… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 32 CITATIONS

Learning Invariances for Policy Generalization

VIEW 5 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

MetaReg: Towards Domain Generalization using Meta-Regularization

VIEW 10 EXCERPTS
CITES RESULTS, METHODS & BACKGROUND
HIGHLY INFLUENCED

Episodic Training for Domain Generalization

VIEW 10 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Generalizable Person Re-Identification by Domain-Invariant Mapping Network

Jifei Song, Yongxin Yang, Yi-Zhe Song, Tao Xiang, Timothy M. Hospedales
  • CVPR 2019
  • 2019
VIEW 7 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Gradient-Based Meta-Learning with Learned Layerwise Metric and Subspace

VIEW 4 EXCERPTS
CITES BACKGROUND & RESULTS
HIGHLY INFLUENCED

References

Publications referenced by this paper.
SHOWING 1-10 OF 35 REFERENCES

Deeper, Broader and Artier Domain Generalization

  • 2017 IEEE International Conference on Computer Vision (ICCV)
  • 2017
VIEW 20 EXCERPTS
HIGHLY INFLUENTIAL

OpenAI Gym

VIEW 3 EXCERPTS
HIGHLY INFLUENTIAL

Domain Generalization for Object Recognition with Multi-task Autoencoders

  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
VIEW 12 EXCERPTS
HIGHLY INFLUENTIAL

Similar Papers

Loading similar papers…