# FAIRLABEL: Correcting Bias in Labels

arXiv: https://arxiv.org/abs/2311.00638 · HTML: https://mlsanity.com/paper/2311.00638/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 30 references that are in the map.
- First posted on arXiv 1 Nov 2023.

## Builds on (most cited first, 13 of 13)

- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [Certifying and removing disparate impact](https://mlsanity.com/paper/1412.3756/) (2014)
- [Identifying and Correcting Label Bias in Machine Learning](https://mlsanity.com/paper/1901.04966/) (2019)
- [The Dataset Nutrition Label: A Framework To Drive Higher Data Quality Standards](https://mlsanity.com/paper/1805.03677/) (2018)
- [Bias in Machine Learning Software: Why? How? What to do?](https://mlsanity.com/paper/2105.12195/) (2021)
- [Achieving non-discrimination in data release](https://mlsanity.com/paper/1611.07438/) (2016)
- [Towards Standardization of Data Licenses: The Montreal Data License](https://mlsanity.com/paper/1903.12262/) (2019)
- [FFPDG: Fast, Fair and Private Data Generation](https://mlsanity.com/paper/2307.00161/) (2023)
- [Using Simpson's Paradox to Discover Interesting Patterns in Behavioral Data](https://mlsanity.com/paper/1805.03094/) (2018)
- [Can you Trust the Trend: Discovering Simpson's Paradoxes in Social Data](https://mlsanity.com/paper/1801.04385/) (2018)
- [Approximate, Adapt, Anonymize (3A): a Framework for Privacy Preserving Training Data Release for Machine Learning](https://mlsanity.com/paper/2307.01875/) (2023)

## Abstract (arXiv)

There are several algorithms for measuring fairness of ML models. A fundamental assumption in these approaches is that the ground truth is fair or unbiased. In real-world datasets, however, the ground truth often contains data that is a result of historical and societal biases and discrimination. Models trained on these datasets will inherit and propagate the biases to the model outputs. We propose FAIRLABEL, an algorithm which detects and corrects biases in labels. The goal of FAIRLABELis to reduce the Disparate Impact (DI) across groups while maintaining high accuracy in predictions. We propose metrics to measure the quality of bias correction and validate FAIRLABEL on synthetic datasets and show that the label correction is correct 86.7% of the time vs. 71.9% for a baseline model. We also apply FAIRLABEL on benchmark datasets such as UCI Adult, German Credit Risk, and Compas datasets and show that the Disparate Impact Ratio increases by as much as 54.2%.