single source · optimization
Distributed AdamW
AdamW is used in a distributed optimization setup, with the cited configuration specifying optimizer hyperparameters and gradient clipping.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See optimizer for the methods filed beside it.
Also called Distributed AdamW optimization, Distributed AdamW optimization for QAD.
Filed under optimization › optimizer