Glossary

Parameter — Glossary

A learned numeric value in a model, such as a neural network weight.

Also known as: model parameter, weights

Reviewed:

Full definition

Parameters are updated during training to reduce loss. Hyperparameters, by contrast, are chosen by practitioners (learning rate, batch size). Parameter counts are a rough capacity signal, not a quality guarantee.

Related encyclopedia topic

Model Parameters — Model parameters are the learned numeric values—weights and biases—that a neural network adjusts during training so it can map inputs to useful outputs.

← Back to glossary · Learning hub