Research

Post-processing for Individual Fairness

NeurIPS

Authors

Published on

10/26/2021

Post-processing in algorithmic fairness is a versatile approach for correcting bias in ML systems that are already used in production. The main appeal of postprocessing is that it avoids expensive retraining. In this work, we propose general post-processing algorithms for individual fairness (IF). We consider a setting where the learner only has access to the predictions of the original model and a similarity graph between individuals, guiding the desired fairness constraints. We cast the IF post-processing problem as a graph smoothing problem corresponding to graph Laplacian regularization that preserves the desired “treat similar individuals similarly” interpretation. Our theoretical results demonstrate the connection of the new objective function to a local relaxation of the original individual fairness. Empirically, our post-processing algorithms correct individual biases in large-scale NLP models such as BERT, while preserving accuracy.

Please cite our work using the BibTeX below.

@misc{https://doi.org/10.48550/arxiv.2110.13796,
  doi = {10.48550/ARXIV.2110.13796},
  
  url = {https://arxiv.org/abs/2110.13796},
  
  author = {Petersen, Felix and Mukherjee, Debarghya and Sun, Yuekai and Yurochkin, Mikhail},
  
  keywords = {Machine Learning (stat.ML), Machine Learning (cs.LG), FOS: Computer and information sciences, FOS: Computer and information sciences},
  
  title = {Post-processing for Individual Fairness},
  
  publisher = {arXiv},
  
  year = {2021},
  
  copyright = {arXiv.org perpetual, non-exclusive license}
}
Close Modal