AI Blog

AI Blog

by Michele Laurelli

Output Layer

/ˈaʊtpʊt ˈleɪər/
Architecture
Definition

The final layer of a neural network that produces predictions or outputs.

Output layer size matches the number of target classes (classification) or output dimensions (regression). Uses appropriate activation: softmax for multi-class, sigmoid for binary, linear for regression.

Examples

1

10 neurons for digit classification

2

1 neuron for binary classification

3

Multiple outputs for regression

Michele Laurelli - AI Research & Engineering