GClasses::GEmpiricalGradientDescent Class Reference

This algorithm does a gradient descent by feeling a small distance out in each dimension to measure the gradient. For efficiency reasons, it only measures the gradient in one dimension (which it cycles round-robin style) per iteration and uses the remembered gradient in the other dimensions. More...

#include <GHillClimber.h>

Inheritance diagram for GClasses::GEmpiricalGradientDescent:

Public Member Functions
	GEmpiricalGradientDescent (GTargetFunction *pCritic, GRand *pRand)
virtual	~GEmpiricalGradientDescent ()
virtual double *	currentVector ()
	Returns the best vector yet found. More...
virtual double	iterate ()
	Performs a little more optimization. (Call this in a loop until acceptable results are found.) More...
void	setLearningRate (double d)
	Sets the learning rate. More...
void	setMomentum (double d)
	Sets the momentum value. More...
Public Member Functions inherited from GClasses::GOptimizer
	GOptimizer (GTargetFunction *pCritic)
virtual	~GOptimizer ()
double	searchUntil (size_t nBurnInIterations, size_t nIterations, double dImprovement)
	This will first call iterate() nBurnInIterations times, then it will repeatedly call iterate() in blocks of nIterations times. If the error heuristic has not improved by the specified ratio after a block of iterations, it will stop. (For example, if the error before the block of iterations was 50, and the error after is 49, then training will stop if dImprovement is > 0.02.) If the error heuristic is not stable, then the value of nIterations should be large. More...

Protected Member Functions
void	reset ()

Protected Attributes
double	m_dFeelDistance
double	m_dLearningRate
double	m_dMomentum
size_t	m_nDimensions
double *	m_pDelta
double *	m_pGradient
GRand *	m_pRand
double *	m_pVector
Protected Attributes inherited from GClasses::GOptimizer
GTargetFunction *	m_pCritic

Detailed Description

This algorithm does a gradient descent by feeling a small distance out in each dimension to measure the gradient. For efficiency reasons, it only measures the gradient in one dimension (which it cycles round-robin style) per iteration and uses the remembered gradient in the other dimensions.

Constructor & Destructor Documentation

GClasses::GEmpiricalGradientDescent::GEmpiricalGradientDescent	(	GTargetFunction *	pCritic,
		GRand *	pRand
	)

virtual GClasses::GEmpiricalGradientDescent::~GEmpiricalGradientDescent ( )

virtual

Member Function Documentation

virtual double* GClasses::GEmpiricalGradientDescent::currentVector ( )

inlinevirtual

Returns the best vector yet found.

Implements GClasses::GOptimizer.

virtual double GClasses::GEmpiricalGradientDescent::iterate ( )

virtual

Performs a little more optimization. (Call this in a loop until acceptable results are found.)

Implements GClasses::GOptimizer.

void GClasses::GEmpiricalGradientDescent::reset ( )

protected

void GClasses::GEmpiricalGradientDescent::setLearningRate ( double  d )

inline

Sets the learning rate.

void GClasses::GEmpiricalGradientDescent::setMomentum ( double  d )

inline

Sets the momentum value.

Member Data Documentation

double GClasses::GEmpiricalGradientDescent::m_dFeelDistance

protected

double GClasses::GEmpiricalGradientDescent::m_dLearningRate

protected

double GClasses::GEmpiricalGradientDescent::m_dMomentum

protected

size_t GClasses::GEmpiricalGradientDescent::m_nDimensions

protected

double* GClasses::GEmpiricalGradientDescent::m_pDelta

protected

double* GClasses::GEmpiricalGradientDescent::m_pGradient

protected

GRand* GClasses::GEmpiricalGradientDescent::m_pRand

protected

double* GClasses::GEmpiricalGradientDescent::m_pVector

protected