Jul 15, 2020 · This paper provides a general framework to analyze the convergence of federated heterogeneous optimization algorithms.
[PDF] Tackling the Objective Inconsistency Problem in Heterogeneous ...
par.nsf.gov › servlets › purl
Using insights from this analysis, we propose FedNova, a normal- ized averaging method that eliminates objective inconsistency while preserving fast error ...
This paper provides a general framework to analyze the convergence of heterogeneous federated optimization algorithms. It subsumes previously proposed methods ...
This paper provides a general framework to analyze the convergence of federated heterogeneous optimization algorithms. It subsumes previously proposed methods ...
The authors study the effect of the clients performing different number of local update steps in federated learning.
Jul 15, 2020 · Using insights from this analysis, we propose Fed-. Nova, a normalized averaging method that eliminates objective inconsistency while preserving ...
This paper provides a general framework to analyze the convergence of federated heterogeneous optimization algorithms.
This paper provides a general framework to analyze the convergence of federated heterogeneous optimization algorithms. It subsumes previously proposed methods ...
The paper studies federated learning, when agents perform different number of local update steps. It shows that normalizing these updates by the effective ...