Deep learning optimizers should be designed to induce desired solution properties, not just convergence speed; different optimizers demonstrably land in qualitatively different minima.
When does preconditioning help or hurt generalization? 2020
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
unclear 1representative citing papers
citing papers explorer
-
Optimizers Qualitatively Alter Solutions And We Should Leverage This
Deep learning optimizers should be designed to induce desired solution properties, not just convergence speed; different optimizers demonstrably land in qualitatively different minima.