# Addressing Discretization-Induced Bias in Demographic Prediction

arXiv: https://arxiv.org/abs/2405.16762 · HTML: https://mlsanity.com/paper/2405.16762/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 87 references that are in the map.
- First posted on arXiv 27 May 2024.

## Builds on (most cited first, 30 of 39)

- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [CVXPY: A Python-Embedded Modeling Language for Convex Optimization](https://mlsanity.com/paper/1603.00943/) (2016)
- [Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints](https://mlsanity.com/paper/1707.09457/) (2017)
- [Fairness in Machine Learning: A Survey](https://mlsanity.com/paper/2010.04053/) (2020)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes](https://mlsanity.com/paper/1711.08412/) (2017)
- [Evaluating Gender Bias in Machine Translation](https://mlsanity.com/paper/1906.00591/) (2019)
- [Women also Snowboard: Overcoming Bias in Captioning Models](https://mlsanity.com/paper/1803.09797/) (2018)
- [Why Is My Classifier Discriminatory?](https://mlsanity.com/paper/1805.12002/) (2018)
- [Towards a Critical Race Methodology in Algorithmic Fairness](https://mlsanity.com/paper/1912.03593/) (2019)
- [Fairness Under Unawareness: Assessing Disparity When Protected Class Is Unobserved](https://mlsanity.com/paper/1811.11154/) (2018)
- [Feedback Loop and Bias Amplification in Recommender Systems](https://mlsanity.com/paper/2007.13019/) (2020)
- [Assessing Algorithmic Fairness with Unobserved Protected Class Using Data Combination](https://mlsanity.com/paper/1906.00285/) (2019)
- [A Systematic Study of Bias Amplification](https://mlsanity.com/paper/2201.11706/) (2022)
- [Data Feedback Loops: Model-driven Amplification of Dataset Biases](https://mlsanity.com/paper/2209.03942/) (2022)
- [Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluation](https://mlsanity.com/paper/2205.04610/) (2022)
- ["What We Can't Measure, We Can't Understand": Challenges to Demographic Data Procurement in the Pursuit of Fairness](https://mlsanity.com/paper/2011.02282/) (2020)
- [Directional Bias Amplification](https://mlsanity.com/paper/2102.12594/) (2021)
- [Who Audits the Auditors? Recommendations from a field scan of the algorithmic auditing ecosystem](https://mlsanity.com/paper/2310.02521/) (2023)
- [Supply-Side Equilibria in Recommender Systems](https://mlsanity.com/paper/2206.13489/) (2022)
- [Feature-Wise Bias Amplification](https://mlsanity.com/paper/1812.08999/) (2018)
- [Active Fairness in Algorithmic Decision Making](https://mlsanity.com/paper/1810.00031/) (2018)
- [Mitigating Gender Bias Amplification in Distribution by Posterior Regularization](https://mlsanity.com/paper/2005.06251/) (2020)
- [Predicting Race and Ethnicity From the Sequence of Characters in a Name](https://mlsanity.com/paper/1805.02109/) (2018)
- [When Fair Ranking Meets Uncertain Inference](https://mlsanity.com/paper/2105.02091/) (2021)
- [Image Cropping on Twitter: Fairness Metrics, their Limitations, and the Importance of Representation, Design, and Agency](https://mlsanity.com/paper/2105.08667/) (2021)
- [Men Also Do Laundry: Multi-Attribute Bias Amplification](https://mlsanity.com/paper/2210.11924/) (2022)
- [Fair Allocation through Selective Information Acquisition](https://mlsanity.com/paper/1911.02715/) (2019)
- [Coarse race data conceals disparities in clinical risk score performance](https://mlsanity.com/paper/2304.09270/) (2023)
- [Improved Bayes Risk Can Yield Reduced Social Welfare Under Competition](https://mlsanity.com/paper/2306.14670/) (2023)

## Built on by (most cited first, 4 of 4)

- [Learning With Multi-Group Guarantees For Clusterable Subpopulations](https://mlsanity.com/paper/2410.14588/) (2024)
- [A Critical Review of Predominant Bias in Neural Networks](https://mlsanity.com/paper/2502.11031/) (2025)
- [Near-Exponential Savings for Mean Estimation with Active Learning](https://mlsanity.com/paper/2511.05736/) (2025)
- [Coarsening Latent-Class Probabilities: Directional Distortion and Coverage Loss](https://mlsanity.com/paper/2608.11784/) (2026)

## Abstract (arXiv)

Racial and other demographic imputation is necessary for many applications, especially in auditing disparities and outreach targeting in political campaigns. The canonical approach is to construct continuous predictions -- e.g., based on name and geography -- and then to textitdiscretize the predictions by selecting the most likely class (argmax). We study how this practice produces textitdiscretization bias. In particular, we show that argmax labeling, as used by a prominent commercial voter file vendor to impute race/ethnicity, results in a substantial under-count of African-American voters, e.g., by 28.2% points in North Carolina. This bias can have substantial implications in downstream tasks that use such labels. We then introduce a textitjoint optimization approach -- and a tractable textitdata-driven thresholding heuristic -- that can eliminate this bias, with negligible individual-level accuracy loss. Finally, we theoretically analyze discretization bias, show that calibrated continuous models are insufficient to eliminate it, and that an approach such as ours is necessary. Broadly, we warn researchers and practitioners against discretizing continuous demographic predictions without considering downstream consequences.