README.md

Doubly robust proximal causal learning for continuous treatments

This repo contains the official implementation for the ICLR2024 paper: Doubly robust proximal causal learning for continuous treatments.

We proposea kernel-based Doubly Robust estimator that can well handle continuous treatments for proximal causal learning.

To reproduce results

Doubly robust datasets

Modify the dataset and sample parameters in the config.py:

python main.py

Citation

@inproceedings{wudoubly,
  title={Doubly Robust Proximal Causal Learning for Continuous Treatments},
  author={Wu, Yong and Fu, Yanwei and Wang, Shouyan and Sun, Xinwei},
  booktitle={The Twelfth International Conference on Learning Representations}
  pages={},
  year={2024}
}

Contact

yongwu21@m.fudan.edu.cn