Multi-objective Optimization for Counterfactual Explanation with Structural Causal Model

Our work pay attention to counterfactual explanation with the structural causal model and multiobjective optimization

Author: Dung Duong, Qian Li, Guandong Xu

This is the code used for the paper Prototype-based Counterfactual Explanation for Causal Classification. I submitted this paper to IJCAI 2021 and got rejected. This work is still in progress. I appreciate your feedback to improve my work. Contact me at TriDung.Duong@student.uts.edu.au

How to run

Train auto-encoder model

python /home/trduong/Data/multiobj-scm-cf/src/ml_cfexplainer/run_algorithm/dfencoder_adult.py
python /home/trduong/Data/multiobj-scm-cf/src/ml_cfexplainer/run_algorithm/dfencoder_adult.py

Reproduce the result

python /multiobj-scm-cf/src/run_simplebn.py
python /multiobj-scm-cf/src/run_adult.py
python /multiobj-scm-cf/src/run_credit.py
python /multiobj-scm-cf/src/run_sangiovese.py

Code Structure

Citing

If you find our work useful for your research work, please cite it as follows.

Reference:

• Mahajan, D., Tan, C., & Sharma, A. (2019). Preserving causal constraints in counterfactual explanations for machine learning classifiers. arXiv preprint arXiv:1912.03277.

• Van Looveren, A., & Klaise, J. (2019). Interpretable counterfactual explanations guided by prototypes. arXiv preprint arXiv:1907.02584.