Stochastic Optimization on Penn Treebank (Character Level) 3x1000 LSTM - 500 Epochs
Metric: Bit per Character (BPC) (higher is better)
Metric: Bit per Character (BPC) (higher is better)
| # | Model↕ | Bit per Character (BPC)▼ | Augmentations | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | AdaBound | 2.863 | No | Domain-independent Dominance of Adaptive Methods | 2019-12-04 | Code |
| 2 | AdaShift | 1.274 | No | Domain-independent Dominance of Adaptive Methods | 2019-12-04 | Code |
| 3 | AdamW | 1.23 | No | Domain-independent Dominance of Adaptive Methods | 2019-12-04 | Code |
| 4 | AvaGrad | 1.175 | No | Domain-independent Dominance of Adaptive Methods | 2019-12-04 | Code |