A Probabilistic Fluctuation based Membership Inference Attack for Diffusion Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Fu, Wenjie, Wang, Huandong, Zhang, Liyuan, Gao, Chen, Li, Yong, Jiang, Tao
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929647217803264
author Fu, Wenjie
Wang, Huandong
Zhang, Liyuan
Gao, Chen
Li, Yong
Jiang, Tao
author_facet Fu, Wenjie
Wang, Huandong
Zhang, Liyuan
Gao, Chen
Li, Yong
Jiang, Tao
contents Membership Inference Attack (MIA) identifies whether a record exists in a machine learning model's training set by querying the model. MIAs on the classic classification models have been well-studied, and recent works have started to explore how to transplant MIA onto generative models. Our investigation indicates that existing MIAs designed for generative models mainly depend on the overfitting in target models. However, overfitting can be avoided by employing various regularization techniques, whereas existing MIAs demonstrate poor performance in practice. Unlike overfitting, memorization is essential for deep learning models to attain optimal performance, making it a more prevalent phenomenon. Memorization in generative models leads to an increasing trend in the probability distribution of generating records around the member record. Therefore, we propose a Probabilistic Fluctuation Assessing Membership Inference Attack (PFAMI), a black-box MIA that infers memberships by detecting these trends via analyzing the overall probabilistic fluctuations around given records. We conduct extensive experiments across multiple generative models and datasets, which demonstrate PFAMI can improve the attack success rate (ASR) by about 27.9% when compared with the best baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2308_12143
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Probabilistic Fluctuation based Membership Inference Attack for Diffusion Models
Fu, Wenjie
Wang, Huandong
Zhang, Liyuan
Gao, Chen
Li, Yong
Jiang, Tao
Machine Learning
Artificial Intelligence
Cryptography and Security
Computer Vision and Pattern Recognition
Membership Inference Attack (MIA) identifies whether a record exists in a machine learning model's training set by querying the model. MIAs on the classic classification models have been well-studied, and recent works have started to explore how to transplant MIA onto generative models. Our investigation indicates that existing MIAs designed for generative models mainly depend on the overfitting in target models. However, overfitting can be avoided by employing various regularization techniques, whereas existing MIAs demonstrate poor performance in practice. Unlike overfitting, memorization is essential for deep learning models to attain optimal performance, making it a more prevalent phenomenon. Memorization in generative models leads to an increasing trend in the probability distribution of generating records around the member record. Therefore, we propose a Probabilistic Fluctuation Assessing Membership Inference Attack (PFAMI), a black-box MIA that infers memberships by detecting these trends via analyzing the overall probabilistic fluctuations around given records. We conduct extensive experiments across multiple generative models and datasets, which demonstrate PFAMI can improve the attack success rate (ASR) by about 27.9% when compared with the best baseline.
title A Probabilistic Fluctuation based Membership Inference Attack for Diffusion Models
topic Machine Learning
Artificial Intelligence
Cryptography and Security
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2308.12143