New analysis framework yields tighter linear convergence for FedExProx on non-strongly convex quadratics and PL functions, proving outperformance over GD once communication costs are counted.
Sharper rates and flexible framework for nonconvex SGD with client and data sampling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Tighter Performance Theory of FedExProx
New analysis framework yields tighter linear convergence for FedExProx on non-strongly convex quadratics and PL functions, proving outperformance over GD once communication costs are counted.