Get Started
Home
Topics
Search
Library
Research questionHow can federated learning protect client updates from inference while resisting Byzantine manipulation?Keeping records on institutional servers does not prevent exchanged updates from revealing information about local data. At the same time, compromised clients can submit harmful updates that distort the shared model.
AI
Machine Learning
Statistical Machine Learning
Technology
Latest papersRecent research connected to this question, newest first.Differentially private federated learning with Byzantine-robust aggregation: A cross-domain framework for secure model training in banking and healthcare systemsApplies to simulated federated classification resembling banking fraud detection and clinical-risk scoring, with 20 clients communicating for 60 rounds and 25% Byzantine participants. The evidence evaluates a Gaussian-mechanism differential privacy layer combined with coordinate-wise trimmed-mean aggregation; it reports privacy loss and classification performance under these settings rather than establishing broad deployment guarantees.research paper · Sep 2, 2026
Related questions
How can federated learning reject poisoned client updates without server validation data or participant history?How can differentially private federated learning preserve faithful explanations without sacrificing predictive utility?How can federated unlearning prevent classifier broadcasts from revealing and enabling reinsertion of deleted client data?How can federated learning protect confidence calibration from attacks that preserve accuracy?