Rescaled ASGD recovers convergence to the true global objective by rescaling worker stepsizes proportional to computation times, matching the known time lower bound in the leading term under non-convex smoothness and bounded heterogeneity.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
PushCen-ADFL is a centroid-based asynchronous federated learning method that applies average-preserving push-sum mixing and regularization to reduce aggregation bias and model drift, claiming up to 6% accuracy gains and 80% lower communication on vision tasks.
citing papers explorer
-
Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity
Rescaled ASGD recovers convergence to the true global objective by rescaling worker stepsizes proportional to computation times, matching the known time lower bound in the leading term under non-convex smoothness and bounded heterogeneity.
-
On the Push-Based Asynchronous Federated Learning: A Bias-Correction Aggregation Approach
PushCen-ADFL is a centroid-based asynchronous federated learning method that applies average-preserving push-sum mixing and regularization to reduce aggregation bias and model drift, claiming up to 6% accuracy gains and 80% lower communication on vision tasks.