Package edu.cmu.tetrad.search.utils

Class GaussianProcessRBF

java.lang.Object
edu.cmu.tetrad.search.utils.GaussianProcessRBF
public class GaussianProcessRBF extends Object
GaussianProcessRBF simulates a Gaussian Process (GP) with a Radial Basis Function (RBF) kernel. It provides functionality for generating simulated function values based on the covariance matrix of the RBF kernel and evaluating the function at arbitrary points using interpolation.

  • Constructor Summary

    Constructors
    Constructor
    Description
    GaussianProcessRBF(double[] xValues, double lengthScale, double amplitude, double noiseStd)
    Constructor for Gaussian Process simulation using an RBF kernel.

  • Method Summary

    Modifier and Type
    Method
    Description
    double
    adjustedEvaluate(double x)
    Evaluates the adjusted value of the simulated function at a given point.
    double
    evaluate(double x)
    Evaluates the simulated function at a given point using interpolation.
    static void
    main(String[] args)
    The main method serves as the entry point for the Gaussian Process simulation demonstration.

    Methods inherited from class java.lang.Object

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

  • Constructor Details

    • GaussianProcessRBF

      public GaussianProcessRBF(double[] xValues, double lengthScale, double amplitude, double noiseStd)
      Constructor for Gaussian Process simulation using an RBF kernel.
      Parameters:
      xValues - Input points for simulation.
      lengthScale - Length scale parameter for the RBF kernel.
      amplitude - Amplitude parameter for the RBF kernel.
      noiseStd - Small noise for numerical stability.

  • Method Details

    • main

      public static void main(String[] args)
      The main method serves as the entry point for the Gaussian Process simulation demonstration. It creates a series of input points, initializes the simulation parameters, and uses those to create and evaluate a Gaussian Process with an RBF kernel.
      Parameters:
      args - Command-line arguments passed to the program, not used in this demonstration.

    • evaluate

      public double evaluate(double x)
      Evaluates the simulated function at a given point using interpolation.
      Parameters:
      x - The input value.
      Returns:
      The interpolated function value.

    • adjustedEvaluate

      public double adjustedEvaluate(double x)
      Evaluates the adjusted value of the simulated function at a given point. The adjustment is done by subtracting the function value at 0.
      Parameters:
      x - The input value at which the adjusted function is evaluated.
      Returns:
      The adjusted function value at the input point.