Hyperparameters

Intermediate

Configuration choices not learned directly (or not typically learned) that govern training or architecture.

Full Definition

Configuration choices not learned directly (or not typically learned) that govern training or architecture.

Keywords

learning rate depth regularization

Domains

Optimization

Related Terms

Objective Function related to

A scalar measure optimized during training, typically expected loss over data, sometimes with regularization terms.

 Empirical Risk Minimization related to

Minimizing average loss on training data; can overfit when data is limited or biased.

 Log Loss related to

Penalizes confident wrong predictions heavily; standard for classification and language modeling.

 Mean Squared Error related to

Average of squared residuals; common regression objective.

Concept Map

See how Hyperparameters connects to other concepts.

Open Knowledge Graph