AMSGrad
E1790152
UNEXPLORED
AMSGrad is an optimization algorithm for training machine learning models that modifies Adam to ensure more stable and theoretically convergent learning by using non-decreasing second-moment estimates.
All labels observed (1)
| Label | Occurrences |
|---|---|
| AMSGrad canonical | 2 |
Referenced by (2)
Full triples — surface form annotated when it differs from this entity's canonical label.