Layer-partitioned Langevin integrators (LOL and AdLaLa) train single hidden layer perceptrons faster, more accurately, and more robustly than SGD/Adam on hard spiral and trigonometric classification problems.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Partitioned integrators for thermodynamic parameterization of neural networks
Layer-partitioned Langevin integrators (LOL and AdLaLa) train single hidden layer perceptrons faster, more accurately, and more robustly than SGD/Adam on hard spiral and trigonometric classification problems.