FairSHAP: Preprocessing for Fairness Through Attribution-Based Data Augmentation

Lin Zhu, Yijun Bian, Lei You
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:2773-2781, 2026.

Abstract

Ensuring fairness in machine learning models is critical, particularly in high-stakes domains where biased decisions can lead to serious societal consequences. However, existing preprocessing approaches generally lack transparent mechanisms for identifying which features are responsible for unfairness. This obscures the rationale behind data modifications. We introduce FairSHAP, a novel preprocessing framework that leverages Shapley value attribution to improve both individual and group fairness. FairSHAP identifies fairness-critical features in the training data using an interpretable measure of feature importance, and systematically modifies them through instance-level matching across sensitive groups. Our method effectively reduces discriminative risk (DR) with an instance-wise guarantee up to an interaction residual term, which is bounded under local matching, while simultaneously bounding the upper limit of demographic parity (DP), which in practice leads to its reduction. Experiments on multiple tabular datasets show that we achieve state-of-the-art or comparable performance across DR, DP, and equality of opportunity (EO) with minimal modifications, thereby preserving data fidelity. As a model-agnostic and transparent method, FairSHAP integrates seamlessly into existing machine learning pipelines and provides actionable insights into the sources of bias. Our code is available on \url{https://github.com/ZhuMuMu0216/FairSHAP.}

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-zhu26b, title = { FairSHAP: Preprocessing for Fairness Through Attribution-Based Data Augmentation }, author = {Zhu, Lin and Bian, Yijun and You, Lei}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {2773--2781}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/zhu26b/zhu26b.pdf}, url = {https://proceedings.mlr.press/v300/zhu26b.html}, abstract = { Ensuring fairness in machine learning models is critical, particularly in high-stakes domains where biased decisions can lead to serious societal consequences. However, existing preprocessing approaches generally lack transparent mechanisms for identifying which features are responsible for unfairness. This obscures the rationale behind data modifications. We introduce FairSHAP, a novel preprocessing framework that leverages Shapley value attribution to improve both individual and group fairness. FairSHAP identifies fairness-critical features in the training data using an interpretable measure of feature importance, and systematically modifies them through instance-level matching across sensitive groups. Our method effectively reduces discriminative risk (DR) with an instance-wise guarantee up to an interaction residual term, which is bounded under local matching, while simultaneously bounding the upper limit of demographic parity (DP), which in practice leads to its reduction. Experiments on multiple tabular datasets show that we achieve state-of-the-art or comparable performance across DR, DP, and equality of opportunity (EO) with minimal modifications, thereby preserving data fidelity. As a model-agnostic and transparent method, FairSHAP integrates seamlessly into existing machine learning pipelines and provides actionable insights into the sources of bias. Our code is available on \url{https://github.com/ZhuMuMu0216/FairSHAP.} } }
Endnote
%0 Conference Paper %T FairSHAP: Preprocessing for Fairness Through Attribution-Based Data Augmentation %A Lin Zhu %A Yijun Bian %A Lei You %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-zhu26b %I PMLR %P 2773--2781 %U https://proceedings.mlr.press/v300/zhu26b.html %V 300 %X Ensuring fairness in machine learning models is critical, particularly in high-stakes domains where biased decisions can lead to serious societal consequences. However, existing preprocessing approaches generally lack transparent mechanisms for identifying which features are responsible for unfairness. This obscures the rationale behind data modifications. We introduce FairSHAP, a novel preprocessing framework that leverages Shapley value attribution to improve both individual and group fairness. FairSHAP identifies fairness-critical features in the training data using an interpretable measure of feature importance, and systematically modifies them through instance-level matching across sensitive groups. Our method effectively reduces discriminative risk (DR) with an instance-wise guarantee up to an interaction residual term, which is bounded under local matching, while simultaneously bounding the upper limit of demographic parity (DP), which in practice leads to its reduction. Experiments on multiple tabular datasets show that we achieve state-of-the-art or comparable performance across DR, DP, and equality of opportunity (EO) with minimal modifications, thereby preserving data fidelity. As a model-agnostic and transparent method, FairSHAP integrates seamlessly into existing machine learning pipelines and provides actionable insights into the sources of bias. Our code is available on \url{https://github.com/ZhuMuMu0216/FairSHAP.}
APA
Zhu, L., Bian, Y. & You, L.. (2026). FairSHAP: Preprocessing for Fairness Through Attribution-Based Data Augmentation . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:2773-2781 Available from https://proceedings.mlr.press/v300/zhu26b.html.

Related Material