Class FullGaussianModel

java.lang.Object

com.knezevic.edaf.v3.models.continuous.FullGaussianModel

All Implemented Interfaces:

Model<RealVector>

public final class FullGaussianModel
extends Object
implements Model<RealVector>

Full-covariance Gaussian model (N(μ, Σ)) with Cholesky sampling.

Model update uses empirical elite estimates blended with optional momentum:

   μ invalid input: '<'- (1-α) μ + α μ_hat
   Σ invalid input: '<'- (1-α) Σ + α Σ_hat
 

followed by optional diagonal shrinkage and symmetric regularization.

Sampling draws z ~ N(0, I) and computes x = μ + Lz, where LLᵀ = Σ.

Constructor Summary

Constructors

Constructor	Description
FullGaussianModel(double jitter)
FullGaussianModel(double jitter, double learningRate, double shrinkage)

Method Summary

Modifier and Type	Method	Description
double[][]	covariance()
ModelDiagnostics	diagnostics()	Exposes diagnostics (entropy, covariance condition number, etc.).
void	fit(List<Individual<RealVector>> selected, Representation<RealVector> representation, RngStream rng)	Fits model parameters from selected individuals.
double[]	mean()
String	name()	Model identifier used in configuration and listing commands.
void	restore(double[] mean, double[][] covariance)	Restores full Gaussian state from checkpoint payload.
List<RealVector>	sample(int count, Representation<RealVector> representation, Problem<RealVector> problem, ConstraintHandling<RealVector> constraintHandling, RngStream rng)	Samples new genotypes from model.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface com.knezevic.edaf.v3.core.api.Model

update

Constructor Details

FullGaussianModel

public FullGaussianModel(double jitter)

Parameters:

jitter - diagonal regularization term added before decomposition

FullGaussianModel

public FullGaussianModel(double jitter,
 double learningRate,
 double shrinkage)

Parameters:

jitter - diagonal regularization term

learningRate - blend factor used for temporal smoothing of mean/covariance

shrinkage - off-diagonal shrinkage factor in [0,1]

Method Details

name

public String name()

Description copied from interface: Model

Model identifier used in configuration and listing commands.

Specified by:

name in interface Model<RealVector>

fit

public void fit(List<Individual<RealVector>> selected,
 Representation<RealVector> representation,
 RngStream rng)

Description copied from interface: Model

Fits model parameters from selected individuals.

Specified by:

fit in interface Model<RealVector>

sample

public List<RealVector> sample(int count,
 Representation<RealVector> representation,
 Problem<RealVector> problem,
 ConstraintHandling<RealVector> constraintHandling,
 RngStream rng)

Description copied from interface: Model

Samples new genotypes from model.

Specified by:

sample in interface Model<RealVector>

diagnostics

public ModelDiagnostics diagnostics()

Description copied from interface: Model

Exposes diagnostics (entropy, covariance condition number, etc.).

Specified by:

diagnostics in interface Model<RealVector>

mean

public double[] mean()

covariance

public double[][] covariance()

restore

public void restore(double[] mean,
 double[][] covariance)

Restores full Gaussian state from checkpoint payload.