L2Regularization<T, TInput, TOutput>

Models & Types · AiDotNet.Regularization

Implements L2 regularization (also known as Ridge), a technique that adds a penalty equal to the square of the magnitude of coefficients.

Inheritance: RegularizationBase<T, TInput, TOutput> → L2Regularization<T, TInput, TOutput>

For Beginners

L2 regularization helps create smoother models by making all coefficients smaller.

Think of it like a gentle pull that shrinks all coefficients proportionally:

  • It doesn’t eliminate features entirely (unlike L1 regularization)
  • It reduces the impact of all features by making their coefficients smaller
  • It particularly penalizes large coefficient values

How It Works

L2 regularization adds a penalty term to the loss function equal to the sum of the squared values of the model coefficients, multiplied by a regularization strength parameter. This encourages smaller, more evenly distributed coefficient values, which helps prevent overfitting and improves model stability.

Constructors

Constructor Summary
L2Regularization(RegularizationOptions options) Initializes a new instance of the L2Regularization class with optional custom options.

Methods

Method Summary
Matrix<T> Regularize(Matrix<T> data) Applies L2 regularization to a vector.
TOutput Regularize(TOutput gradient, TOutput coefficients)
Vector<T> Regularize(Vector<T> data) Applies L2 regularization to a vector.
Vector<T> Regularize(Vector<T> gradient, Vector<T> coefficients)