Experiments indicate that small-batch SGD promotes flatter loss minima and better generalization in overparameterized networks, and that sparse subnetworks can retain nearly full accuracy.
Deep double descent: Where bigger models and more data can hurt
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Implicit Regularization and Generalization in Overparameterized Neural Networks
Experiments indicate that small-batch SGD promotes flatter loss minima and better generalization in overparameterized networks, and that sparse subnetworks can retain nearly full accuracy.