# A Fairness Analysis on Private Aggregation of Teacher Ensembles

arXiv: https://arxiv.org/abs/2109.08630 · HTML: https://mlsanity.com/paper/2109.08630/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 31 references that are in the map.
- First posted on arXiv 17 Sep 2021.

## Builds on (most cited first, 16 of 16)

- [Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks](https://mlsanity.com/paper/1511.04508/) (2015)
- [Renyi Differential Privacy](https://mlsanity.com/paper/1702.07476/) (2017)
- [Differential Privacy Has Disparate Impact on Model Accuracy](https://mlsanity.com/paper/1905.12101/) (2019)
- [Scalable Private Learning with PATE](https://mlsanity.com/paper/1802.08908/) (2018)
- [Fairness without Demographics through Adversarially Reweighted Learning](https://mlsanity.com/paper/2006.13114/) (2020)
- [Differentially Private Empirical Risk Minimization Revisited: Faster and More General](https://mlsanity.com/paper/1802.05251/) (2018)
- [Differentially Private Fair Learning](https://mlsanity.com/paper/1812.02696/) (2018)
- [Training individually fair ML models with Sensitive Subspace Robustness](https://mlsanity.com/paper/1907.00020/) (2019)
- [Efficient Private ERM for Smooth Objectives](https://mlsanity.com/paper/1703.09947/) (2017)
- [Neither Private Nor Fair: Impact of Data Imbalance on Utility and Fairness in Differential Privacy](https://mlsanity.com/paper/2009.06389/) (2020)
- [Fair Decision Making using Privacy-Protected Data](https://mlsanity.com/paper/1905.12744/) (2019)
- [Differentially Private and Fair Deep Learning: A Lagrangian Dual Approach](https://mlsanity.com/paper/2009.12562/) (2020)
- [Fair Learning with Private Demographic Data](https://mlsanity.com/paper/2002.11651/) (2020)
- [DP-SGD vs PATE: Which Has Less Disparate Impact on Model Accuracy?](https://mlsanity.com/paper/2106.12576/) (2021)
- [Decision Making with Differential Privacy under a Fairness Lens](https://mlsanity.com/paper/2105.07513/) (2021)
- [AI-SARAH: Adaptive and Implicit Stochastic Recursive Gradient Methods](https://mlsanity.com/paper/2102.09700/) (2021)

## Built on by (most cited first, 11 of 11)

- [Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Data](https://mlsanity.com/paper/2109.11429/) (2021)
- [Differential Privacy and Fairness in Decisions and Learning Tasks: A Survey](https://mlsanity.com/paper/2202.08187/) (2022)
- [Learning with Impartiality to Walk on the Pareto Frontier of Fairness, Privacy, and Utility](https://mlsanity.com/paper/2302.09183/) (2023)
- [On the Alignment of Group Fairness with Attribute Privacy](https://mlsanity.com/paper/2211.10209/) (2022)
- [An Empirical Analysis of Fairness Notions under Differential Privacy](https://mlsanity.com/paper/2302.02910/) (2023)
- [SF-PATE: Scalable, Fair, and Private Aggregation of Teacher Ensembles](https://mlsanity.com/paper/2204.05157/) (2022)
- [DP-SGD vs PATE: Which Has Less Disparate Impact on GANs?](https://mlsanity.com/paper/2111.13617/) (2021)
- [SoK: Unintended Interactions among Machine Learning Defenses and Risks](https://mlsanity.com/paper/2312.04542/) (2023)
- [On the Fairness Impacts of Private Ensembles Models](https://mlsanity.com/paper/2305.11807/) (2023)
- [SoK: Taming the Triangle -- On the Interplays between Fairness, Interpretability and Privacy in Machine Learning](https://mlsanity.com/paper/2312.16191/) (2023)
- [PRISM: Diversifying Dataset Distillation by Decoupling Architectural Priors](https://mlsanity.com/paper/2511.09905/) (2025)

## Abstract (arXiv)

The Private Aggregation of Teacher Ensembles (PATE) is an important private machine learning framework. It combines multiple learning models used as teachers for a student model that learns to predict an output chosen by noisy voting among the teachers. The resulting model satisfies differential privacy and has been shown effective in learning high-quality private models in semisupervised settings or when one wishes to protect the data labels. This paper asks whether this privacy-preserving framework introduces or exacerbates bias and unfairness and shows that PATE can introduce accuracy disparity among individuals and groups of individuals. The paper analyzes which algorithmic and data properties are responsible for the disproportionate impacts, why these aspects are affecting different groups disproportionately, and proposes guidelines to mitigate these effects. The proposed approach is evaluated on several datasets and settings.