A consensus-based distributed stochastic gradient algorithm is shown to converge in mean square to a critical point for non-convex problems, and it trains distributed neural networks on MNIST with accuracy comparable to centralized training.
Communication-efficient learning of deep networks from de centralized data,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Distributed Stochastic Gradient Method for Non-Convex Problems with Applications in Supervised Learning
A consensus-based distributed stochastic gradient algorithm is shown to converge in mean square to a critical point for non-convex problems, and it trains distributed neural networks on MNIST with accuracy comparable to centralized training.