Large-batch methods converge to sharp minima causing a generalization gap, while small-batch methods reach flat minima due to inherent gradient noise.
Such and algorithm might also improve training time through faster convergence
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.LG 1years
2016 1verdicts
UNVERDICTED 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
Large-batch methods converge to sharp minima causing a generalization gap, while small-batch methods reach flat minima due to inherent gradient noise.