(ICML 2024) Feature attribution with necessity and sufficiency via dual-stage perturbation test for causal explanation 🔥

This is a PyTorch implementation of paper:

Feature attribution with necessity and sufficiency via dual-stage perturbation test for causal explanation

Authors:

Xuexin Chen, Ruichu Cai, Zhengting Huang, Yuxuan Zhu, Julien Horwood, Zhifeng Hao, Zijian Li, Jose Miguel Hernandez-Lobato

How to run our code

run the following script: ./main.py

Citation

If you find this code useful, please cite the following:

@inproceedings{chen2024feature,
  title={Feature Attribution with Necessity and Sufficiency via Dual-stage Perturbation Test for Causal Explanation},
  author={Chen, Xuexin and Cai, Ruichu and Huang, Zhengting and Zhu, Yuxuan and Horwood, Julien and Hao, Zhifeng and Li, Zijian and Hern{\'a}ndez-Lobato, Jos{\'e} Miguel},
  booktitle={International Conference on Machine Learning},
  pages={6486--6502},
  year={2024},
  organization={PMLR}
}