AdamW

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology4 results

Stochastic OptimizationonCIFAR-10 WRN-28-10 - 200 Epochs
Accuracy· 2019-12-04
95.89
best: 96.36 (Adam (eps-adjusted))
Domain-independent Dominance of Adaptive Methods arXiv:1912.01823
Stochastic OptimizationonImageNet ResNet-50 - 90 Epochs
Top 1 Accuracy· 2019-12-04
72.9
best: 76.51 (AvaGrad)
Domain-independent Dominance of Adaptive Methods arXiv:1912.01823
Stochastic OptimizationonPenn Treebank (Character Level) 3x1000 LSTM - 500 Epochs
Bit per Character (BPC)· 2019-12-04
1.23
best: 2.863 (AdaBound)
Domain-independent Dominance of Adaptive Methods arXiv:1912.01823
Stochastic OptimizationonCIFAR-100 WRN-28-10 - 200 Epochs
Accuracy· 2019-12-04
79.87
best: 81.24 (AvaGrad)
Domain-independent Dominance of Adaptive Methods arXiv:1912.01823