single source · optimization
Gradient accumulation
Accumulates gradients across training computations as a lever for scalable distributed training.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See optimization for the methods filed beside it.
Filed under optimization