<ul><li>Federated Learning (FL) can be affected by data and device heterogeneities, caused by clients' different local data distributions and latencies in uploading model updates.</li><li>Traditional FL schemes consider these heterogeneities as separate aspects, but in practical scenarios, they are intertwined.</li><li>A new FL framework is presented in this paper, which converts stale model updates into unstale ones to tackle intertwined heterogeneities.</li><li>The approach estimates the distributions of clients' local training data from stale model updates and improves model accuracy by up to 25%.</li></ul>

Tackling Intertwined Data and Device Heterogeneities in Federated Learning with Unlimited Staleness

Discover more