Adversarial Task Up-sampling for Meta-learning

NeurIPS'22: Adversarial Task Up-sampling for Meta-learning (Official Pytorch implementation).

Data

Related datasets are in Google Drive

Setups

The requiring environment is as bellow:

• Linux
• Python 3+
• PyTorch 1.7.1

If you find this repository useful in your research, please cite the following paper:

@inproceedings{wu2022adversarial,
  title={Adversarial Task Up-sampling for Meta-learning},
  author={Wu, Yichen and Huang, Long-Kai and Wei, Ying},
  booktitle={Advances in Neural Information Processing Systems},
  year={2022}
}

Acknowledgements

We thank the Pytorch implementation of Metamix(https://github.com/huaxiuyao/MetaMix).

Contact: Yichen Wu (wuyichen.am97@gmail.com).