# Provably Fair Representations

arXiv: https://arxiv.org/abs/1710.04394 · HTML: https://mlsanity.com/paper/1710.04394/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Oct 2017.

## Built on by (most cited first, 22 of 22)

- [Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey](https://mlsanity.com/paper/2207.07068/) (2022)
- [A Clarification of the Nuances in the Fairness Metrics Landscape](https://mlsanity.com/paper/2106.00467/) (2021)
- [Controllable Guarantees for Fair Outcomes via Contrastive Information Estimation](https://mlsanity.com/paper/2101.04108/) (2021)
- [Individual Fairness Guarantees for Neural Networks](https://mlsanity.com/paper/2205.05763/) (2022)
- [FARE: Provably Fair Representation Learning with Practical Certificates](https://mlsanity.com/paper/2210.07213/) (2022)
- [Removing biased data to improve fairness and accuracy](https://mlsanity.com/paper/2102.03054/) (2021)
- [Fair Representation: Guaranteeing Approximate Multiple Group Fairness for Unknown Tasks](https://mlsanity.com/paper/2109.00545/) (2021)
- [Input-agnostic Certified Group Fairness via Gaussian Parameter Smoothing](https://mlsanity.com/paper/2206.11423/) (2022)
- [Fair Representations by Compression](https://mlsanity.com/paper/2105.14044/) (2021)
- [Private and Reliable Neural Network Inference](https://mlsanity.com/paper/2210.15614/) (2022)
- [10 Years of Fair Representations: Challenges and Opportunities](https://mlsanity.com/paper/2407.03834/) (2024)
- [Fair Group-Shared Representations with Normalizing Flows](https://mlsanity.com/paper/2201.06336/) (2022)
- [Fair Interpretable Representation Learning with Correction Vectors](https://mlsanity.com/paper/2202.03078/) (2022)
- [Fair-CDA: Continuous and Directional Augmentation for Group Fairness](https://mlsanity.com/paper/2304.00295/) (2023)
- [Causal Fair Metric: Bridging Causality, Individual Fairness, and Adversarial Robustness](https://mlsanity.com/paper/2310.19391/) (2023)
- [Fair Text Classification via Transferable Representations](https://mlsanity.com/paper/2503.07691/) (2025)
- [From Global to Local: Social Bias Transfer in CLIP](https://mlsanity.com/paper/2508.17750/) (2025)
- [Fair Interpretable Learning via Correction Vectors](https://mlsanity.com/paper/2201.06343/) (2022)
- [Learning Fair Representations with High-Confidence Guarantees](https://mlsanity.com/paper/2310.15358/) (2023)
- [Computability, Complexity, Consistency and Controllability: A Four C's Framework for cross-disciplinary Ethical Algorithm Research](https://mlsanity.com/paper/2102.04234/) (2021)
- [Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference](https://mlsanity.com/paper/2510.21017/) (2025)
- [Measuring and Mitigating Bias in Code Generated by Large Language Models](https://mlsanity.com/paper/2606.00049/) (2026)

## Abstract (arXiv)

Machine learning systems are increasingly used to make decisions about people's lives, such as whether to give someone a loan or whether to interview someone for a job. This has led to considerable interest in making such machine learning systems fair. One approach is to transform the input data used by the algorithm. This can be achieved by passing each input data point through a representation function prior to its use in training or testing. Techniques for learning such representation functions from data have been successful empirically, but typically lack theoretical fairness guarantees. We show that it is possible to prove that a representation function is fair according to common measures of both group and individual fairness, as well as useful with respect to a target task. These provable properties can be used in a governance model involving a data producer, a data user and a data regulator, where there is a separation of concerns between fairness and target task utility to ensure transparency and prevent perverse incentives. We formally define the 'cost of mistrust' of using this model compared to the setting where there is a single trusted party, and provide bounds on this cost in particular cases. We present a practical approach to learning fair representation functions and apply it to financial and criminal justice datasets. We evaluate the fairness and utility of these representation functions using measures motivated by our theoretical results.