Source code of the paper Improving Generalization in Meta-learning via Task Augmentation. This code is built upon the pytorch implementation of few-shot learning few-shot and the implementation of MAML++, MR-MAML.

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

@inproceedings{yao2021improving,
  title={Improving Generalization in Meta-learning via Task Augmentation},
  author={Yao, Huaxiu and Huang, Longkai and Zhang, Linjun and Wei, Ying and Tian, Li and Zou, James and Huang, Junzhou and Li, Zhenhui},
  booktitle={Proceeding of the Thirty-eighth International Conference on Machine Learning},
  year={2021} 
}

We have put the related datasets in Google Drive

• python 3.*
• Pytorch 1.17+
• Tensorflow 1.15 (for Pose only)

Please see the bash file in /Drug for hyperparameter settings

Please see the bash file in /Pose for hyperparameter settings

Please see the bash file in /miniImagenet for hyperparameter settings

Please see the bash file in /omniglot for hyperparameter settings