# Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping

arXiv: https://arxiv.org/abs/2506.01396 · HTML: https://mlsanity.com/paper/2506.01396/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 43 references that are in the map.
- First posted on arXiv 2 Jun 2025.

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Algorithmic decision making and the cost of fairness](https://mlsanity.com/paper/1701.08230/) (2017)
- [Opacus: User-Friendly Differential Privacy Library in PyTorch](https://mlsanity.com/paper/2109.12298/) (2021)
- [Differential Privacy Has Disparate Impact on Model Accuracy](https://mlsanity.com/paper/1905.12101/) (2019)
- [Unlocking High-Accuracy Differentially Private Image Classification through Scale](https://mlsanity.com/paper/2204.13650/) (2022)
- [Differentially Private Learning with Adaptive Clipping](https://mlsanity.com/paper/1905.03871/) (2019)
- [Hyperparameter Tuning with Renyi Differential Privacy](https://mlsanity.com/paper/2110.03620/) (2021)
- [Automatic Clipping: Differentially Private Deep Learning Made Easier and Stronger](https://mlsanity.com/paper/2206.07136/) (2022)
- [Private Selection from Private Candidates](https://mlsanity.com/paper/1811.07971/) (2018)
- [Differential Privacy and Fairness in Decisions and Learning Tasks: A Survey](https://mlsanity.com/paper/2202.08187/) (2022)
- [Disparate Impact in Differential Privacy from Gradient Misalignment](https://mlsanity.com/paper/2206.07737/) (2022)
- [The Trade-Offs of Private Prediction](https://mlsanity.com/paper/2007.05089/) (2020)
- [Practical Differentially Private Hyperparameter Tuning with Subsampling](https://mlsanity.com/paper/2301.11989/) (2023)
- [On (assessing) the fairness of risk score models](https://mlsanity.com/paper/2302.08851/) (2023)
- [On the Convergence of DP-SGD with Adaptive Clipping](https://mlsanity.com/paper/2412.19916/) (2024)
- [FairDP: Certified Fairness with Differential Privacy](https://mlsanity.com/paper/2305.16474/) (2023)

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

- [When Gradient Clipping Becomes a Control Mechanism for Differential Privacy in Deep Learning](https://mlsanity.com/paper/2602.10584/) (2026)
- [Differential Privacy in Two-Layer Networks: How DP-SGD Harms Fairness and Robustness](https://mlsanity.com/paper/2603.04881/) (2026)

## Abstract (arXiv)

Differential privacy (DP) has become an essential framework for privacy-preserving machine learning. Existing DP learning methods, however, often have disparate impacts on model predictions, e.g., for minority groups. Gradient clipping, which is often used in DP learning, can suppress larger gradients from challenging samples. We show that this problem is amplified by adaptive clipping, which will often shrink the clipping bound to tiny values to match a well-fitting majority, while significantly reducing the accuracy for others. We propose bounded adaptive clipping, which introduces a tunable lower bound to prevent excessive gradient suppression. Our method improves worst-class accuracy by over 10 percentage points on Skewed and Fashion MNIST compared to unbounded adaptive clipping, 7 points compared to Automatic clipping, and 5 points compared to constant clipping. The code is available at https://github.com/TrustworthyMLHelsinki/adaptive-clipping-fairness.