paper-with-me

홈 › Papers

SRATTA : Sample Re-ATTribution Attack of Secure Aggregation in Federated Learning

2023-06-13 · Tanguy Marchand, Régis Loeb, Ulysse Marteau-Ferey, Jean Ogier du Terrail, Arthur Pignet

We consider a cross-silo federated learning (FL) setting where a machine learning model with a fully connected first layer is trained between different clients and a central server using FedAvg, and where the aggregation step can be performed with secure aggregation (SA). We present SRATTA an attack relying only on aggregated models which, under realistic assumptions, (i) recovers data samples from the different clients, and (ii) groups data samples coming from the same client together. While sample recovery has already been explored in an FL setting, the ability to group samples per client, despite the use of SA, is novel. This poses a significant unforeseen security threat to FL and effectively breaks SA. We show that SRATTA is both theoretically grounded and can be used in practice on realistic models and datasets. We also propose counter-measures, and claim that clients should play an active role to guarantee their privacy during training.

📄 PDF Abstract BibTeX arXiv:2306.07644

Code (1)

owkin/sratta 공식 구현 pytorch

Tasks

Federated Learning

Similar Papers 제목 키워드 기반

Eluding Secure Aggregation in Federated Learning via Model Inconsistency

2021-11-14 · Dario Pasquini, Danilo Francati, Giuseppe Ateniese

Secure aggregation is a cryptographic protocol that securely computes the aggregation of its inputs. It is pivotal in keeping model updates private in federated learning. Indeed, the use of secure aggregation prevents th…

Federated Learning

Gradient Disaggregation: Breaking Privacy in Federated Learning by Reconstructing the User Participant Matrix

2021-06-10 · Maximilian Lam, Gu-Yeon Wei, David Brooks, Vijay Janapa Reddi 외

We show that aggregated model updates in federated learning may be insecure. An untrusted central server may disaggregate user updates from sums of updates across participants given repeated observations, enabling the se…

Federated Learning

LOKI: Large-scale Data Reconstruction Attack against Federated Learning through Model Manipulation

2023-03-21 · Joshua C. Zhao, Atul Sharma, Ahmed Roushdy Elkordy, Yahya H. Ezzeldin 외

Federated learning was introduced to enable machine learning over large decentralized datasets while promising privacy by eliminating the need for data sharing. Despite this, prior work has shown that shared gradients of…

Federated LearningReconstruction Attack

Uncovering Attacks and Defenses in Secure Aggregation for Federated Deep Learning

2024-10-13 · Yiwei Zhang, Rouzbeh Behnia, Attila A. Yavuz, Reza Ebrahimi 외

Federated learning enables the collaborative learning of a global model on diverse data, preserving data locality and eliminating the need to transfer user data to a central server. However, data privacy remains vulnerab…

Federated Learning

CachePrune: Neural-Based Attribution Defense Against Indirect Prompt Injection Attacks

2025-04-29 · Rui Wang, Junda Wu, Yu Xia, Tong Yu 외

Large Language Models (LLMs) are identified as being susceptible to indirect prompt injection attack, where the model undesirably deviates from user-provided instructions by executing tasks injected in the prompt context…

Instruction Following