Concept
Local Response Normalization
Local Response Normalization (LRN) normalizes each activation by dividing it by a factor computed from the activations of neighboring channels at the same spatial position. It comes from the AlexNet era, where it was used after certain ReLU layers to encourage competition…
The rest of “Local Response Normalization” 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→