Namespace NeuralNetworks.Regularization

Classes

BaseRegularizer

Provides a base implementation of IRegularizer for applying regularization updates and optionally reporting a regularization penalty in the loss.

CompositeRegularizer

Combines multiple regularizers into a single regularization pipeline.

ElasticNetRegularizer

Combines L1 and L2 regularization into a single elastic net regularizer.

L1Regularizer

Applies L1 regularization by adding a penalty proportional to the sum of absolute parameter values.

L2Regularizer

Applies L2 regularization by adding a penalty proportional to the squared magnitude of model parameters.

L3Regularizer

Applies L3 regularization by adding a penalty proportional to the cubed magnitude of model parameters.

MaxNormRegularizer

Applies max-norm regularization by scaling a parameter vector when its L2 norm exceeds a specified limit.

NullRegularizer

Represents a regularizer that applies no regularization.

Interfaces

IRegularizer

Defines regularization operations for applying gradient-based adjustments, reporting penalty terms, and updating parameters after optimizer steps.