Beginner · Neural Networks

Layer

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. A group of neurons that perform the same kind of transformation in parallel.

Technical Definition

A group of neurons that perform the same kind of transformation in parallel.

How it works

Networks are stacks of layers: fully-connected (dense), convolutional, recurrent, attention, normalization, dropout. Each layer transforms its input tensor and passes the result to the next. Depth (number of layers) is a key driver of representational power.

Related Concepts

• Hidden Layer — Any layer in a neural network that is neither the input nor the output.
• Input Layer — The first layer of a neural network, where raw features enter the model.
• Output Layer — The final layer of a neural network that produces the prediction.