# Post-processing for Fair Regression via Explainable SVD

arXiv: https://arxiv.org/abs/2504.03093 · HTML: https://mlsanity.com/paper/2504.03093/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 20 references that are in the map.
- First posted on arXiv 4 Apr 2025.

## Builds on (most cited first, 3 of 3)

- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Conscientious Classification: A Data Scientist's Guide to Discrimination-Aware Classification](https://mlsanity.com/paper/1907.09013/) (2019)
- [Achieving Equalized Odds by Resampling Sensitive Attributes](https://mlsanity.com/paper/2006.04292/) (2020)

## Built on by (most cited first, 2 of 2)

- [Demographic-Agnostic Fairness without Harm](https://mlsanity.com/paper/2509.24077/) (2025)
- [Individual Fairness In Strategic Classification](https://mlsanity.com/paper/2602.05084/) (2026)

## Abstract (arXiv)

This paper presents a post-processing algorithm for training fair neural network regression models that satisfy statistical parity, utilizing an explainable singular value decomposition (SVD) of the weight matrix. We propose a linear transformation of the weight matrix, whereby the singular values derived from the SVD of the transformed matrix directly correspond to the differences in the first and second moments of the output distributions across two groups. Consequently, we can convert the fairness constraints into constraints on the singular values. We analytically solve the problem of finding the optimal weights under these constraints. Experimental validation on various datasets demonstrates that our method achieves a similar or superior fairness-accuracy trade-off compared to the baselines without using the sensitive attribute at the inference time.