SWB-DM: A Calibrated Sliced-Wasserstein-Barycenter Aggregator with Delayed-Momentum Caching for Byzantine-Robust Federated Learning under Partial Participation
SWB-DM's cache carries a real warm-up cost, but extending all baselines to the same round budget shows its CIFAR-10 gains are disproportionately large, and on CIFAR-100, FLTrust benefits more -- for reasons entirely unrelated to caching.
S. Saranraj, S. SaranyaM, S. AlexDavid et al.
· 0 citations