Inference Attack

Inference attacks exploit machine learning model outputs to infer sensitive information about the training data, posing a significant privacy risk in various applications like federated learning and AI-as-a-service. Current research focuses on developing novel attack techniques targeting different model architectures (e.g., graph neural networks, large language models) and data modalities, as well as designing robust defenses such as differential privacy and adversarial training. Understanding and mitigating these attacks is crucial for ensuring the responsible deployment of machine learning systems and protecting user privacy in collaborative and cloud-based settings.

Papers

August 11, 2022

Shielding Federated Learning Systems against Inference Attacks with ARM TrustZone
Aghiles Ait Messaoud, Sonia Ben Mokhtar, Vlad Nitu, Valerio Schiavoni
Inference Attack Novel Attack Trusted Execution Environment Protection Mechanism

August 9, 2022

Adversarial Machine Learning-Based Anticipation of Threats Against Vehicle-to-Microgrid Services
Ahmed Omara, Burak Kantarci
Adversarial Learning White Box Service Provider Inference Attack Threat Word Gray Box Vehicle to Grid

August 4, 2022

Differentially Private Counterfactuals via Functional Mechanism
Fan Yang, Qizhang Feng, Kaixiong Zhou, Jiahao Chen, Xia Hu
Differential Privacy Counterfactual Explanation High Quality Counterfactuals Functional Mechanism Inference Attack

July 24, 2022

Privacy Against Inference Attacks in Vertical Federated Learning
Borzoo Rassouli, Morteza Varasteh, Deniz Gunduz
Privacy Policy Sensitive Data Vertical Federated Learning Inference Attack Reconstruction Attack Passive Party

June 3, 2022

On the Privacy Properties of GAN-generated Samples
Zinan Lin, Vyas Sekar, Giulia Fanti
Generative Adversarial Network GAN Model External Sample Membership Inference Inference Attack

May 13, 2022

Collaborative Drug Discovery: Inference-level Data Protection Perspective
Balazs Pejo, Mina Remeli, Adam Arany, Mathieu Galtier, Gergely Acs
Drug Discovery Privacy Risk Inference Attack Model Collaboration Drug Candidate Secure Inference

April 22, 2022

Federated Learning: Balancing the Thin Line Between Data Intelligence and Privacy
Sherin Mary Mathews, Samuel A. Assefa
Privacy Policy Inference Attack Model Poisoning Attack Security Challenge Business Intelligence Fragmented Personal Data

March 31, 2022

Truth Serum: Poisoning Machine Learning Models to Reveal Their Secrets
Florian Tramèr, Reza Shokri, Ayrton San Joaquin, Hoang Le, Matthew Jagielski, Sanghyun Hong, Nicholas Carlini
Scientific Inference Membership Inference Attack Privacy Guarantee Inference Attack Untargeted Attack Dark Secret

March 13, 2022

One Parameter Defense -- Defending against Data Inference Attacks via Differential Privacy
Dayong Ye, Sheng Shen, Tianqing Zhu, Bo Liu, Wanlei Zhou
Differential Privacy Membership Inference Attack Model Inversion Attack Inference Attack Differential Privacy Mechanism Parameter Based Attack

February 15, 2022

OLIVE: Oblivious Federated Learning on Trusted Execution Environment against the risk of sparsification
Fumiyuki Kato, Yang Cao, Masatoshi Yoshikawa
Federated Learning High Quality Risk Description Secure Aggregation Random Sparsification Inference Attack Privacy Preserving Federated Learning Trusted Execution Environment Olive Oil Manufacturing

February 10, 2022

PPA: Preference Profiling Attack Against Federated Learning
Chunyi Zhou, Yansong Gao, Anmin Fu, Kai Chen, Zhiyang Dai, Zhi Zhang, Minhui Xue, Yuqing Zhang
PRIvacy Leakage Inference Attack Privacy Inference Preference Poisoning

February 7, 2022

Redactor: A Data-centric and Individualized Defense Against Inference Attacks
Geon Heo, Steven Euijong Whang
Data Centric Private Data Decision Boundary Data Leakage Inference Attack Data Redaction

February 4, 2022

Dikaios: Privacy Auditing of Algorithmic Fairness via Attribute Inference Attacks
Jan Aalmoes, Vasisht Duddu, Antoine Boutet
Algorithmic Fairness Inference Attack Empirical Privacy Attribute Inference Attack Fairness Audit Attribute Privacy