/Variational-HyperAdam

PyTorch implementation of VR-HyperAdam from our TPAMI paper " Variational-HyperAdam: A Meta-learning Appproach to Network Training"

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

Variational-HyperAdam

This is a PyTorch implementation of the Variational-HyperAdam algorithm from our TPAMI paper:

Title: Variational-HyperAdam: A Meta-learning Appproach to Network Training

Authors: Shipeng Wang, Yan Yang, Jian Sun, Zongben Xu

Email: wangshipeng8128@stu.xjtu.edu.cn; wangshipeng8128@gmail.com

Institution: School of Mathematics and Statistics, Xi'an Jiaotong University

Link: https://ieeexplore.ieee.org/document/9361276

Usage

To replicate the experiments,run from terminal:

cd HyperAdam
sh batch_process.sh

Requirement: PyTorch >= 1.0, Python 3.7

Citation

If the code is useful in your research, please cite ourpaper:

@ARTICLE{vrhyperadam2021wang,
  author={S. {Wang} and Y. {Yang} and J. {Sun} and Z. {Xu}},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
  title={Variational HyperAdam: A Meta-learning Approach to Network Training}, 
  year={2021},
  volume={},
  number={},
  pages={1-1},
  doi={10.1109/TPAMI.2021.3061581}
  }