2026
Towards Explainable Privacy Preservation in Federated Learning via Shapley Value-Guided Noise Injection
ICASSP 2026poster
This paper proposes FedSVA, an explainable differential privacy (DP) mechanism for federated learning (FL) that dynamically calibrates noise injection based on the privacy contribution of attributes via Shapley Values. Unlike heuristic DP methods, FedSVA quantifies each attribute's influence on mode…