# Removing biased data to improve fairness and accuracy

arXiv: https://arxiv.org/abs/2102.03054 · HTML: https://mlsanity.com/paper/2102.03054/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 49 of its 128 references that are in the map.
- First posted on arXiv 5 Feb 2021.

## Builds on (most cited first, 30 of 49)

- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Certifying and removing disparate impact](https://mlsanity.com/paper/1412.3756/) (2014)
- [Fair prediction with disparate impact: A study of bias in recidivism prediction instruments](https://mlsanity.com/paper/1610.07524/) (2016)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [A Reductions Approach to Fair Classification](https://mlsanity.com/paper/1803.02453/) (2018)
- [Inherent Trade-Offs in the Fair Determination of Risk Scores](https://mlsanity.com/paper/1609.05807/) (2016)
- [On Fairness and Calibration](https://mlsanity.com/paper/1709.02012/) (2017)
- [Algorithmic decision making and the cost of fairness](https://mlsanity.com/paper/1701.08230/) (2017)
- [Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment](https://mlsanity.com/paper/1610.08452/) (2016)
- [AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias](https://mlsanity.com/paper/1810.01943/) (2018)
- [Empirical Risk Minimization under Fairness Constraints](https://mlsanity.com/paper/1802.08626/) (2018)
- [The Variational Fair Autoencoder](https://mlsanity.com/paper/1511.00830/) (2015)
- [A comparative study of fairness-enhancing interventions in machine learning](https://mlsanity.com/paper/1802.04422/) (2018)
- [Avoiding Discrimination through Causal Reasoning](https://mlsanity.com/paper/1706.02744/) (2017)
- [Censoring Representations with an Adversary](https://mlsanity.com/paper/1511.05897/) (2015)
- [Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations](https://mlsanity.com/paper/1707.00075/) (2017)
- [Towards Fairer Datasets: Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchy](https://mlsanity.com/paper/1912.07726/) (2019)
- [A Convex Framework for Fair Regression](https://mlsanity.com/paper/1706.02409/) (2017)
- [Classification with Fairness Constraints: A Meta-Algorithm with Provable Guarantees](https://mlsanity.com/paper/1806.06055/) (2018)
- [Fair Inference On Outcomes](https://mlsanity.com/paper/1705.10378/) (2017)
- [Fairness Testing: Testing Software for Discrimination](https://mlsanity.com/paper/1709.03221/) (2017)
- [Assessing Algorithmic Fairness with Unobserved Protected Class Using Data Combination](https://mlsanity.com/paper/1906.00285/) (2019)
- [Automated Directed Fairness Testing](https://mlsanity.com/paper/1807.00468/) (2018)
- [Achieving Fairness through Adversarial Learning: an Application to Recidivism Prediction](https://mlsanity.com/paper/1807.00199/) (2018)
- [A Confidence-Based Approach for Balancing Fairness and Accuracy](https://mlsanity.com/paper/1601.05764/) (2016)
- [Learning Certified Individually Fair Representations](https://mlsanity.com/paper/2002.10312/) (2020)
- [Operationalizing Individual Fairness with Pairwise Fair Representations](https://mlsanity.com/paper/1907.01439/) (2019)

## Built on by (most cited first, 17 of 17)

- [ZeroCap: Zero-Shot Image-to-Text Generation for Visual-Semantic Arithmetic](https://mlsanity.com/paper/2111.14447/) (2021)
- [Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey](https://mlsanity.com/paper/2207.07068/) (2022)
- [Detecting Shortcut Learning for Fair Medical AI using Shortcut Testing](https://mlsanity.com/paper/2207.10384/) (2022)
- [Accurate Fairness: Improving Individual Fairness without Trading Accuracy](https://mlsanity.com/paper/2205.08704/) (2022)
- [FAIR-FATE: Fair Federated Learning with Momentum](https://mlsanity.com/paper/2209.13678/) (2022)
- [Enforcing Delayed-Impact Fairness Guarantees](https://mlsanity.com/paper/2208.11744/) (2022)
- [Understanding Unfairness in Fraud Detection through Model and Data Bias Interactions](https://mlsanity.com/paper/2207.06273/) (2022)
- [A call for embodied AI](https://mlsanity.com/paper/2402.03824/) (2024)
- [Provable Fairness for Neural Network Models using Formal Verification](https://mlsanity.com/paper/2212.08578/) (2022)
- [Towards Fairness and Privacy: A Novel Data Pre-processing Optimization Framework for Non-binary Protected Attributes](https://mlsanity.com/paper/2410.00836/) (2024)
- [Outlier Detection using AI: A Survey](https://mlsanity.com/paper/2112.00588/) (2021)
- [Repairing Brain-Computer Interfaces with Fault-Based Data Acquisition](https://mlsanity.com/paper/2203.10677/) (2022)
- [Trusting Fair Data: Leveraging Quality in Fairness-Driven Data Removal Techniques](https://mlsanity.com/paper/2405.12926/) (2024)
- [Can Ensembling Pre-processing Algorithms Lead to Better Machine Learning Fairness?](https://mlsanity.com/paper/2212.02614/) (2022)
- [De-Biasing Models of Biased Decisions: A Comparison of Methods Using Mortgage Application Data](https://mlsanity.com/paper/2405.00910/) (2024)
- [The Illusion of Fairness: Auditing Fairness Interventions with Audit Studies](https://mlsanity.com/paper/2507.02152/) (2025)
- [A Blended Likelihood Approach for Achieving Fairness Using Naive Bayes](https://mlsanity.com/paper/2605.25228/) (2026)

## Abstract (arXiv)

Machine learning systems are often trained using data collected from historical decisions. If past decisions were biased, then automated systems that learn from historical data will also be biased. We propose a black-box approach to identify and remove biased training data. Machine learning models trained on such debiased data (a subset of the original training data) have low individual discrimination, often 0%. These models also have greater accuracy and lower statistical disparity than models trained on the full historical data. We evaluated our methodology in experiments using 6 real-world datasets. Our approach outperformed seven previous approaches in terms of individual discrimination and accuracy.