Membership Privacy

Membership privacy in machine learning focuses on protecting the confidentiality of data used to train models, preventing attackers from inferring whether a specific data point was part of the training set. Current research investigates the vulnerability of various architectures, including deep neural networks (both traditional and spiking), and diffusion models, to membership inference attacks, exploring factors like representation magnitude and model parameters' influence on attack success. This field is crucial for responsible AI development, impacting data security and user privacy across diverse applications, from image recognition to healthcare, by quantifying and mitigating the risk of data leakage.

Papers

November 2, 2024

WaKA: Data Attribution using K-Nearest Neighbors and Membership Privacy Principles
Patrick Mesana, Clément Bénesse, Hadrien Lautraite, Gilles Caporossi, Sébastien Gambs
Membership Inference Attack General Principle Inference Attack K Nearest Neighbor Membership Privacy Data Attribution Likelihood Ratio Attack

September 28, 2024

Membership Privacy Evaluation in Deep Spiking Neural Networks
Jiaxin Li, Gorka Abad, Stjepan Picek, Mauro Conti
Neural Network Spiking Neural Network Neuromorphic Datasets Membership Privacy Deep Spiking Neural Network

September 10, 2024

Catch Me if You Can: Detecting Unauthorized Data Use in Deep Learning Models
Zitao Chen, Karthik Pattabiraman
Deep Learning Deep Learning Model Data Provenance Membership Inference Membership Privacy

July 23, 2024

Representation Magnitude has a Liability to Privacy Vulnerability
Xingli Fang, Jung-Eun Kim
Machine Learning Machine Learning Model Privacy Risk Membership Privacy Liability Inducing Text

July 2, 2024

A Method to Facilitate Membership Inference Attacks in Deep Learning Models
Zitao Chen, Karthik Pattabiraman
Deep Learning Model Practical Method Membership Inference Attack Empirical Privacy Membership Privacy

June 17, 2024

Do Parameters Reveal More than Loss for Membership Inference?
Anshuman Suri, Xiao Zhang, David Evans
Membership Inference Attack Many Parameter White Box Membership Inference Interest Loss Membership Privacy Hessian Inverse

February 15, 2024

How Much Does Each Datapoint Leak Your Privacy? Quantifying the Per-datum Membership Leakage
Achraf Azize, Debabrota Basu
Membership Inference Attack Privacy Policy Adaptive Adversary Field Data Membership Privacy Mathematical Proof

February 7, 2024

Impact of Dataset Properties on Membership Inference Vulnerability of Deep Transfer Learning
Marlon Tobaben, Hibiki Ito, Joonas Jälkö, Gauri Pradhan, Yuan He, Antti Honkela
Deep Learning Global Impact Membership Inference Attack Membership Inference Efficient Classification Membership Privacy Privacy Threat Dataset Level

December 7, 2023

Diffence: Fencing Membership Privacy With Diffusion Models
Yuefeng Peng, Ali Naseh, Amir Houmansadr
Diffusion Model Membership Inference Attack State of the Art Defense Membership Privacy Prior Defense

October 5, 2023

Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning
Harsh Chaudhari, Giorgio Severi, Alina Oprea, Jonathan Ullman
Membership Inference Attack Membership Inference Pace Adaptive Membership Privacy Label Attack Poisoning Efficiency Label Only Membership Inference Attack

July 4, 2023

Overconfidence is a Dangerous Thing: Mitigating Membership Inference Attacks by Enforcing Less Confident Prediction
Zitao Chen, Karthik Pattabiraman
Membership Inference Attack Privacy Risk Good Thing Privacy Protection Privacy Utility Trade Membership Privacy Overconfidence Problem

June 14, 2023

A Unified Framework of Graph Information Bottleneck for Robustness and Membership Privacy
Enyan Dai, Limeng Cui, Zhengyang Wang, Xianfeng Tang, Yinghan Wang, Monica Cheng, Bing Yin, Suhang Wang
Graph Neural Network Adversarial Attack Native Robustness Membership Inference Attack Graph Information Bottleneck Membership Privacy

June 8, 2023

Investigating the Effect of Misalignment on Membership Privacy in the White-box Setting
Ana-Maria Cretu, Daniel Jones, Yves-Alexandre de Montjoye, Shruti Tople
Mixed Effect White Box Membership Privacy Segment Misalignment Shadow Model Box Membership Inference Attack Local Ultimate Gradient Inspection Model Misalignment

June 5, 2023

Discriminative Adversarial Privacy: Balancing Accuracy and Membership Privacy in Neural Networks
Eugenio Lomurno, Alberto Archetti, Francesca Ausonio, Matteo Matteucci
Neural Network Differential Privacy Adversarial Training Membership Inference Attack Membership Privacy Adversarial Privacy

February 15, 2023

Data Forensics in Diffusion Models: A Systematic Analysis of Membership Privacy
Derui Zhu, Dingfan Chen, Jens Grossklags, Mario Fritz
Diffusion Model Membership Inference Attack Attack Vector Membership Privacy Digital Forensic

August 23, 2022

Auditing Membership Leakages of Multi-Exit Networks
Zheng Li, Yiyong Liu, Xinlei He, Ning Yu, Michael Backes, Yang Zhang
Membership Inference Attack Attack Performance Multi Exit Membership Privacy Hybrid Attack

March 17, 2022

Leveraging Adversarial Examples to Quantify Membership Information Leakage
Ganesh Del Grosso, Hamid Jalalzai, Georg Pichler, Catuscia Palamidessi, Pablo Piantanida
Training Data Adversarial Example Membership Inference Training Machine Learning Model Membership Privacy

March 10, 2022

Membership Privacy Protection for Image Translation Models via Adversarial Knowledge Distillation
Saeed Ranjbar Alvar, Lanjun Wang, Jian Pei, Yong Zhang
Membership Inference Attack Image to Image Training Model Membership Privacy Image Translation Model Adversarial Knowledge Distillation

December 4, 2021

SHAPr: An Efficient and Versatile Membership Privacy Risk Metric for Machine Learning
Vasisht Duddu, Sebastian Szyller, N. Asokan
Machine Learning Training Data Membership Inference Attack Membership Privacy