A new concentration lemma for almost super-martingales yields a stopping-time convergence rate of O(log(tau)/sqrt(tau)) for a momentum SGD variant, removing one logarithmic factor.
Stochastic first order methods in smooth convex optimization
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
math.OC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Breaking a Logarithmic Barrier in the Stopping Time Convergence Rate of Stochastic First-order Methods
A new concentration lemma for almost super-martingales yields a stopping-time convergence rate of O(log(tau)/sqrt(tau)) for a momentum SGD variant, removing one logarithmic factor.