Empirical study of the Privacy–Robustness–Performance trilemma in Federated Learning: combining DP-SGD, FLTrust Byzantine-robust aggregation, and Top-k compression across 8 configurations on MNIST, simulated with Flower.
machine-learning pytorch mnist flower differential-privacy federated-learning topk byzantine-fault-tolerance privacy-preserving-ml dp-sgd opacus trilemma robust-aggregation fl-trust
-
Updated
Jul 16, 2026 - Python