Concept
Internal covariate shift
Internal covariate shift refers to the change in the distribution of a layer's inputs during training, caused by the continuous updates to all the layers below it. Think of it like training a model on data whose distribution keeps drifting, except the drift is internal:…
The rest of “Internal covariate shift” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→