Class LevenbergMarquardt

java.lang.Object

org.episteme.core.mathematics.optimization.LevenbergMarquardt

public class LevenbergMarquardt
extends Object

Levenberg-Marquardt algorithm for nonlinear least squares.

Combines Gauss-Newton and gradient descent for robust curve fitting.

*

Reference:
Levenberg, K. (1944). A method for the solution of certain non-linear problems in least squares. Quarterly of Applied Mathematics, 2, 164-168.

Since:

1.0

Author:

Silvere Martin-Michiellot, Gemini AI (Google DeepMind)

Constructor Summary

Constructors

Constructor	Description
LevenbergMarquardt()
LevenbergMarquardt(double tolerance, int maxIterations)

Method Summary

Modifier and Type	Method	Description
double[]	fit(BiFunction<double[],Double,Double> model, double[] xData, double[] yData, double[] params0)	Fit a model to data using Levenberg-Marquardt.
double[]	fitExponential(double[] xData, double[] yData)	Fit exponential: y = a * exp(b * x)
double[]	fitGaussian(double[] xData, double[] yData)	Fit Gaussian: y = a * exp(-((x-b)^2) / (2*c^2))
double[]	fitPowerLaw(double[] xData, double[] yData)	Fit power law: y = a * x^b

Methods inherited from class Object

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

Constructor Details

LevenbergMarquardt

public LevenbergMarquardt(double tolerance,
 int maxIterations)

LevenbergMarquardt

public LevenbergMarquardt()

Method Details

fit

public double[] fit(BiFunction<double[],Double,Double> model,
 double[] xData,
 double[] yData,
 double[] params0)

Fit a model to data using Levenberg-Marquardt.

Parameters:

model - Function (params, x) -> predicted y

xData - Independent variable data points

yData - Observed values

params0 - Initial parameter guess

Returns:

Optimized parameters

fitExponential

public double[] fitExponential(double[] xData,
 double[] yData)

Fit exponential: y = a * exp(b * x)

fitGaussian

public double[] fitGaussian(double[] xData,
 double[] yData)

Fit Gaussian: y = a * exp(-((x-b)^2) / (2*c^2))

fitPowerLaw

public double[] fitPowerLaw(double[] xData,
 double[] yData)

Fit power law: y = a * x^b