# Evaluating the Impact of Local Differential Privacy on Utility Loss via Influence Functions

arXiv: https://arxiv.org/abs/2309.08678 · HTML: https://mlsanity.com/paper/2309.08678/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 36 references that are in the map.
- First posted on arXiv 15 Sep 2023.

## Builds on (most cited first, 19 of 19)

- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach](https://mlsanity.com/paper/1609.03683/) (2016)
- [Estimating Training Data Influence by Tracing Gradient Descent](https://mlsanity.com/paper/2002.08484/) (2020)
- [Differentially Private Federated Learning: A Client Level Perspective](https://mlsanity.com/paper/1712.07557/) (2017)
- [RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response](https://mlsanity.com/paper/1407.6981/) (2014)
- [Collecting Telemetry Data Privately](https://mlsanity.com/paper/1712.01524/) (2017)
- [On the Accuracy of Influence Functions for Measuring Group Effects](https://mlsanity.com/paper/1905.13289/) (2019)
- [If Influence Functions are the Answer, Then What is the Question?](https://mlsanity.com/paper/2209.05364/) (2022)
- [On Second-Order Group Influence Functions for Black-Box Predictions](https://mlsanity.com/paper/1911.00418/) (2019)
- [Locally Private Graph Neural Networks](https://mlsanity.com/paper/2006.05535/) (2020)
- [Achieving Fairness at No Utility Cost via Data Reweighing with Influence](https://mlsanity.com/paper/2202.00787/) (2022)
- [Local, Private, Efficient Protocols for Succinct Histograms](https://mlsanity.com/paper/1504.04686/) (2015)
- [Differential Privacy: An Economic Method for Choosing Epsilon](https://mlsanity.com/paper/1402.3329/) (2014)
- [Understanding Instance-Level Impact of Fairness Constraints](https://mlsanity.com/paper/2206.15437/) (2022)
- [Toward Understanding the Influence of Individual Clients in Federated Learning](https://mlsanity.com/paper/2012.10936/) (2020)
- [Local Differential Privacy: a tutorial](https://mlsanity.com/paper/1907.11908/) (2019)
- [Out-of-Distribution Generalization Analysis via Influence Function](https://mlsanity.com/paper/2101.08521/) (2021)
- [Interpreting Robust Optimization via Adversarial Influence Functions](https://mlsanity.com/paper/2010.01247/) (2020)
- [Issues Encountered Deploying Differential Privacy](https://mlsanity.com/paper/1809.02201/) (2018)

## Built on by (most cited first, 1 of 1)

- [Exposing the Vulnerability of Decentralized Learning to Membership Inference Attacks Through the Lens of Graph Mixing](https://mlsanity.com/paper/2412.12837/) (2024)

## Abstract (arXiv)

How to properly set the privacy parameter in differential privacy (DP) has been an open question in DP research since it was first proposed in 2006. In this work, we demonstrate the ability of influence functions to offer insight into how a specific privacy parameter value will affect a model's test loss in the randomized response-based local DP setting. Our proposed method allows a data curator to select the privacy parameter best aligned with their allowed privacy-utility trade-off without requiring heavy computation such as extensive model retraining and data privatization. We consider multiple common randomization scenarios, such as performing randomized response over the features, and/or over the labels, as well as the more complex case of applying a class-dependent label noise correction method to offset the noise incurred by randomization. Further, we provide a detailed discussion over the computational complexity of our proposed approach inclusive of an empirical analysis. Through empirical evaluations we show that for both binary and multi-class settings, influence functions are able to approximate the true change in test loss that occurs when randomized response is applied over features and/or labels with small mean absolute error, especially in cases where noise correction methods are applied.