Noether Networks: meta-learning useful conserved quantities

This repository contains the code necessary to reproduce experiments from "Noether Networks: meta-learning useful conserved quantities." Noether Networks meta-learn inductive biases in the form of useful conserved quantities. For details on the method, check out our NeurIPS 2021 paper, linked on our project website.

For instructions on how to train and evaluate a Noether Network for video prediction, check out video_prediction/README.md.

Citation

If this work is useful to you, please cite our paper:

@inproceedings{
alet2021noether,
title={Noether Networks: meta-learning useful conserved quantities},
author={Ferran Alet and Dylan Doblar and Allan Zhou and Joshua B. Tenenbaum and Kenji Kawaguchi and Chelsea Finn},
booktitle={Thirty-Fifth Conference on Neural Information Processing Systems},
year={2021},
url={https://openreview.net/forum?id=_NOwVKCmSo}
}