A pytorch implementation of Fair Representation Learning through Implicit Path Alignment.

We consider learning a fair representation through learning invariant optimal predictor, on the top of predefined representation. We further fromulate the problem as a bi-level optimization and formally justify the learned representation satisfying the sufficiency rule.

• Pytorch >=1.0,
• Scikit-learn >= 0.19.1

@article{shui2022fair,
  title={Fair Representation Learning through Implicit Path Alignment},
  author={Shui, Changjian and Chen, Qi and Li, Jiaqi and Wang, Boyu and Gagn{\'e}, Christian},
  journal={arXiv preprint arXiv:2205.13316},
  year={2022}
}