TV-HSGT couples hybrid variance-reduced stochastic gradients with AB gradient tracking to obtain a dynamic regret bound over time-varying directed networks without assuming bounded gradients.
Optimization and learning with information streams: Time-varying algorithms and applications
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
A Hybrid Stochastic Gradient Tracking Method for Distributed Online Optimization Over Time-Varying Directed Networks
TV-HSGT couples hybrid variance-reduced stochastic gradients with AB gradient tracking to obtain a dynamic regret bound over time-varying directed networks without assuming bounded gradients.