Concept
Mini-batch
A mini-batch is a small subset of the training set processed in a single forward and backward pass, sitting between two extremes: full-batch gradient descent (one update per epoch over all data) and stochastic gradient descent (one update per example). Instead of computing…
The rest of “Mini-batch” 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→