Causal Structural Hypothesis Testing and Data Generation Models

Download

• Paper + Video
• Code

Abstract

We introduce CSHTEST and CSVHTEST, novel architectures for causal model hypothesis testing and data generation. These models use non-parametric structural causal knowledge and approximate a causal model’s functional relationships using deep neural networks. The architectures are tested on extensive simulated DAGs, a synthetic pendulum dataset, and a real-world medical trauma dataset to demonstrate practical use for causal inference.

Published at the NeurIPS 2022 Workshop on Synthetic Data for Empowering ML Research.

Citation

Bhat, Sunay Gajanan, Omead Pooladzandi, Jeffrey Jiang, and Gregory Pottie. 2022. “Causal Structural Hypothesis Testing and Data Generation Models.” Proceedings of the NeurIPS 2022 Workshop on Synthetic Data for Empowering ML Research.

@inproceedings{bhat2022cshtest,
  title={Causal Structural Hypothesis Testing and Data Generation Models},
  author={Bhat, Sunay Gajanan and Pooladzandi, Omead and Jiang, Jeffrey and Pottie, Gregory},
  booktitle={NeurIPS 2022 Workshop on Synthetic Data for Empowering ML Research},
  year={2022}
}