Why Does Differential Privacy with Large Epsilon Defend Against Practical Membership Inference Attacks?

• URL: http://arxiv.org/abs/2402.09540v1
• Date: Wed, 14 Feb 2024 19:31:45 GMT
• Title: Why Does Differential Privacy with Large Epsilon Defend Against Practical Membership Inference Attacks?
• Authors: Andrew Lowy, Zhuohang Li, Jing Liu, Toshiaki Koike-Akino, Kieran Parsons, Ye Wang
• Abstract summary: For small privacy parameter $epsilon$, $epsilon$-differential privacy (DP) provides a strong worst-case guarantee. Existing DP theory cannot explain empirical findings. We propose a new privacy notion called practical membership privacy (PMP)
• Score: 19.21246519924815
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: For small privacy parameter $\epsilon$, $\epsilon$-differential privacy (DP) provides a strong worst-case guarantee that no membership inference attack (MIA) can succeed at determining whether a person's data was used to train a machine learning model. The guarantee of DP is worst-case because: a) it holds even if the attacker already knows the records of all but one person in the data set; and b) it holds uniformly over all data sets. In practical applications, such a worst-case guarantee may be overkill: practical attackers may lack exact knowledge of (nearly all of) the private data, and our data set might be easier to defend, in some sense, than the worst-case data set. Such considerations have motivated the industrial deployment of DP models with large privacy parameter (e.g. $\epsilon \geq 7$), and it has been observed empirically that DP with large $\epsilon$ can successfully defend against state-of-the-art MIAs. Existing DP theory cannot explain these empirical findings: e.g., the theoretical privacy guarantees of $\epsilon \geq 7$ are essentially vacuous. In this paper, we aim to close this gap between theory and practice and understand why a large DP parameter can prevent practical MIAs. To tackle this problem, we propose a new privacy notion called practical membership privacy (PMP). PMP models a practical attacker's uncertainty about the contents of the private data. The PMP parameter has a natural interpretation in terms of the success rate of a practical MIA on a given data set. We quantitatively analyze the PMP parameter of two fundamental DP mechanisms: the exponential mechanism and Gaussian mechanism. Our analysis reveals that a large DP parameter often translates into a much smaller PMP parameter, which guarantees strong privacy against practical MIAs. Using our findings, we offer principled guidance for practitioners in choosing the DP parameter.

Related papers

• Privacy Amplification for the Gaussian Mechanism via Bounded Support [64.86780616066575]
  Data-dependent privacy accounting frameworks such as per-instance differential privacy (pDP) and Fisher information loss (FIL) confer fine-grained privacy guarantees for individuals in a fixed training dataset. We propose simple modifications of the Gaussian mechanism with bounded support, showing that they amplify privacy guarantees under data-dependent accounting.
  arXiv  Detail & Related papers  (2024-03-07T21:22:07Z)
• Closed-Form Bounds for DP-SGD against Record-level Inference [18.85865832127335]
  We focus on the popular DP-SGD algorithm, and derive simple closed-form bounds. We obtain bounds for membership inference that match state-of-the-art techniques. We present a novel data-dependent bound against attribute inference.
  arXiv  Detail & Related papers  (2024-02-22T09:26:16Z)
• What Are the Chances? Explaining the Epsilon Parameter in Differential Privacy [17.201862983773662]
  Differential privacy (DP) is a mathematical privacy notion increasingly deployed across government and industry. We develop three methods to convey probabilistic DP guarantees to end users. We find that odds-based explanation methods are more effective than output-based methods.
  arXiv  Detail & Related papers  (2023-03-01T18:53:25Z)
• Provable Membership Inference Privacy [31.08016816475564]
  Differential privacy (DP) has emerged as one canonical standard for provable privacy. We propose a novel privacy notion, membership inference privacy (MIP), to address these challenges. We show MIP can be achieved using less amount of randomness compared to the amount required for guaranteeing DP, leading to a smaller drop in utility.
  arXiv  Detail & Related papers  (2022-11-12T06:13:00Z)
• Analyzing Privacy Leakage in Machine Learning via Multiple Hypothesis Testing: A Lesson From Fano [83.5933307263932]
  We study data reconstruction attacks for discrete data and analyze it under the framework of hypothesis testing. We show that if the underlying private data takes values from a set of size $M$, then the target privacy parameter $epsilon$ can be $O(log M)$ before the adversary gains significant inferential power.
  arXiv  Detail & Related papers  (2022-10-24T23:50:12Z)
• Individual Privacy Accounting for Differentially Private Stochastic Gradient Descent [69.14164921515949]
  We characterize privacy guarantees for individual examples when releasing models trained by DP-SGD. We find that most examples enjoy stronger privacy guarantees than the worst-case bound. This implies groups that are underserved in terms of model utility simultaneously experience weaker privacy guarantees.
  arXiv  Detail & Related papers  (2022-06-06T13:49:37Z)
• Large Scale Transfer Learning for Differentially Private Image Classification [51.10365553035979]
  Differential Privacy (DP) provides a formal framework for training machine learning models with individual example level privacy. Private training using DP-SGD protects against leakage by injecting noise into individual example gradients. While this result is quite appealing, the computational cost of training large-scale models with DP-SGD is substantially higher than non-private training.
  arXiv  Detail & Related papers  (2022-05-06T01:22:20Z)
• Optimal Membership Inference Bounds for Adaptive Composition of Sampled Gaussian Mechanisms [93.44378960676897]
  Given a trained model and a data sample, membership-inference (MI) attacks predict whether the sample was in the model's training set. A common countermeasure against MI attacks is to utilize differential privacy (DP) during model training to mask the presence of individual examples. In this paper, we derive bounds for the textitadvantage of an adversary mounting a MI attack, and demonstrate tightness for the widely-used Gaussian mechanism.
  arXiv  Detail & Related papers  (2022-04-12T22:36:56Z)
• Bounding Membership Inference [28.64031194463754]
  We provide a tighter bound on the accuracy of any MI adversary when a training algorithm provides $epsilon$-DP. Our scheme enables $epsilon$-DP users to employ looser DP guarantees when training their model to limit the success of any MI adversary.
  arXiv  Detail & Related papers  (2022-02-24T17:54:15Z)
• On the Practicality of Differential Privacy in Federated Learning by Tuning Iteration Times [51.61278695776151]
  Federated Learning (FL) is well known for its privacy protection when training machine learning models among distributed clients collaboratively. Recent studies have pointed out that the naive FL is susceptible to gradient leakage attacks. Differential Privacy (DP) emerges as a promising countermeasure to defend against gradient leakage attacks.
  arXiv  Detail & Related papers  (2021-01-11T19:43:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.