Statements (22)
Predicate | Object |
---|---|
gptkbp:instanceOf |
mathematical optimization
|
gptkbp:arXivID |
1711.05101
|
gptkbp:category |
stochastic optimization
|
gptkbp:decouples |
weight decay
|
https://www.w3.org/2000/01/rdf-schema#label |
AdamW
|
gptkbp:improves |
gptkb:Adam
|
gptkbp:introduced |
gptkb:Frank_Hutter
Ilya Loshchilov |
gptkbp:introducedIn |
2017
|
gptkbp:parameter |
learning rate
beta1 beta2 epsilon weight decay |
gptkbp:popularFor |
gptkb:TensorFlow
gptkb:PyTorch |
gptkbp:publishedIn |
Decoupled Weight Decay Regularization
|
gptkbp:solvedBy |
L2 regularization coupling in Adam
|
gptkbp:usedFor |
training neural networks
|
gptkbp:usedIn |
deep learning
|
gptkbp:bfsParent |
gptkb:Adaptive_Moment_Estimation
|
gptkbp:bfsLayer |
6
|