adam2392 / causal-component-analysis

Causal Component Analysis (CauCA)

Overview

Causal Component Analysis is a project that bridges the gap between Independent Component Analysis (ICA) and Causal Representation Learning (CRL). This project includes implementations and experiments related to the papers:

1. Liang, W., Kekić, A., von Kügelgen, J., Buchholz, S., Besserve, M., Gresele, L., & Schölkopf, B. (2023). Causal Component Analysis. In Proceedings of the Thirty-seventh Conference on Neural Information Processing Systems.

   The corresponding experiments are in the experiments/cauca folder.

2. von Kügelgen, J., Besserve, M., Liang, W., Gresele, L., Kekić, A., Bareinboim, E., Blei, D., & Schölkopf, B. (2023). Nonparametric Identifiability of Causal Representations from Unknown Interventions. In Proceedings of the Thirty-seventh Conference on Neural Information Processing Systems.

   The corresponding experiments are in the experiments/nonparam_ident folder.

Installation

Clone the repository

git clone [e-mail]:akekic/causal-component-analysis.git

and install the package

pip install -e .

License

This project is licensed under the MIT license. See the LICENSE file for details.

Citation

If you use CauCA, please cite the corresponding paper as follows.

Liang, W., Kekić, A., von Kügelgen, J., Buchholz, S., Besserve, M., Gresele, L., & Schölkopf, B. (2023). Causal Component Analysis. In Proceedings of the Thirty-seventh Conference on Neural Information Processing Systems.

Bibtex

@inproceedings{
    liang2023causal,
    title={Causal Component Analysis},
    author={Wendong Liang and Armin Keki{\'c} and Julius von K{\"u}gelgen and Simon Buchholz and Michel Besserve and Luigi Gresele and Bernhard Sch{\"o}lkopf},
    booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
    year={2023},
    url={https://openreview.net/forum?id=HszLRiHyfO}
}