Skip to content

juntang-zhuang/TorchDiffEqPack

Repository files navigation

PyTorch implementation of ACA and MALI, a reverse accurate and memory efficient solver for Neural ODEs, achieving new SOTA results on image classification, continuous generative modeling, and time-series analysis with Neural ODEs. Please see https://jzkay12.github.io/TorchDiffEqPack/ for detailed documentation and APIs.

If you find this package helpful, please cite our paper:

@inproceedings{
zhuang2021mali,
title={{\{}MALI{\}}: A memory efficient and reverse accurate integrator for Neural {\{}ODE{\}}s},
author={Juntang Zhuang and Nicha C Dvornek and sekhar tatikonda and James s Duncan},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=blfSjHeFM_e}
} 

@inproceedings{zhuang2020adaptive,
  title={Adaptive checkpoint adjoint method for gradient estimation in neural ODE},
  author={Zhuang, Juntang and Dvornek, Nicha and Li, Xiaoxiao and Tatikonda, Sekhar and Papademetris, Xenophon and Duncan, James},
  booktitle={International Conference on Machine Learning},
  pages={11639--11649},
  year={2020},
  organization={PMLR}
}