Research Article
AdaCN: An Adaptive Cubic Newton Method for Nonconvex Stochastic Optimization
Table 1
Summaries of the settings used in experiments.
| Task | Dataset | Network type | Architecture | Optimizer |
| Image classification | Mnist | Convolutional neural network | LeNet | AdaCN, Apollo, Adam, SGD, AdaBound | CIFAR10 CIFAR100 | VGG11, ResNet34, DenseNet121 | Language modeling | Penn Treebank | Recurrent | 1, 2, 3-layer LSTM |
|
|