SAM solves a min-max problem to locate flat low-loss regions, improving generalization on CIFAR, ImageNet and label-noise tasks.
We train all models with weight decay 1e−5 as suggested in (Tan & Le, 2019), but we reduce the learning rate to 0.016 as the models tend to diverge for higher values
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2020 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sharpness-Aware Minimization for Efficiently Improving Generalization
SAM solves a min-max problem to locate flat low-loss regions, improving generalization on CIFAR, ImageNet and label-noise tasks.