3 papers
cs.CR2021
Federated Learning Attacks Revisited: A Critical Discussion of Gaps, Assumptions, and Evaluation Setups
Aidmar Wainakh, Ephraim Zimmer, Sandeep Subedi +5
Federated learning (FL) enables a set of entities to collaboratively train a machine learning model without sharing their sensitive data, thus, mitigating some privacy concerns. Ho…
cs.CR2021
User-Level Label Leakage from Gradients in Federated Learning
Aidmar Wainakh, Fabrizio Ventola, Till Müßig +6
Federated learning enables multiple users to build a joint model by sharing their model updates (gradients), while their raw data remains local on their devices. In contrast to the…
cs.CR2020
Enhancing Privacy via Hierarchical Federated Learning
Aidmar Wainakh, Alejandro Sanchez Guinea, Tim Grube +1
Federated learning suffers from several privacy-related issues that expose the participants to various threats. A number of these issues are aggravated by the centralized architect…