Wavelet Policy: Lifting Scheme for Policy Learning in Long-Horizon Tasks

This is the official repository of Wavelet Policy: Lifting Scheme for Policy Learning in Long-Horizon Tasks.

Environment setup

• Python 3.9+
• PyTorch 1.13.0
• Cuda 11.7

Or install dependencies with conda:

conda env create -f wavelet_policy.yaml

# NOTE: The versions of the dependencies listed above are only for reference. 
If you encounter any errors during the installation through .yaml, you can install manually with 'pip install ...'

Kitchen and CARLA experiments

cd kitchen_carla
# Follow the README.md in that folder.

PushT and Transport experiments

cd pusht_transport
# Follow the README.md in that folder.

D3IL experiments

cd d3il
# Follow the README.md in that folder.

Acknowledgements

Our code is built upon the repositories: bet, diffusion_policy, and d3il. We would appreciate the authors for their great work.

Citation

If you found this repository is helpful, please cite:

@article{huang2025wavelet,
  title={Wavelet Policy: Lifting Scheme for Policy Learning in Long-Horizon Tasks},
  author={Huang, Hao and Yuan, Shuaihang and Bethala, Geeta Chandra Raju and Wen, Congcong and Tzes, Anthony and Fang, Yi},
  journal={arXiv preprint arXiv:2507.04331},
  year={2025}
}