#include <itkFiniteDifferenceGradientDescentOptimizer.h>

Detailed Description

An optimizer based on gradient descent ...

If $C(x)$ is a costfunction that has to be minimised, the following iterative algorithm is used to find the optimal parameters x:

$x(k+1)_j = x(k)_j - a(k) \left[ C(x(k)_j + c(k)) - C(x(k)_j - c(k)) \right] / 2c(k),$

for all parameters $j$ .

From this equation it is clear that it a gradient descent optimizer, using a finite difference approximation of the gradient.

The gain $a(k)$ at each iteration $k$ is defined by:

$a(k) = a / (A + k + 1)^{\alpha}.$

The perturbation size $c(k)$ at each iteration $k$ is defined by:

$c(k) = c / (k + 1)^{\gamma}.$

Note the similarities to the SimultaneousPerturbation optimizer and the StandardGradientDescent optimizer.

See also: FiniteDifferenceGradientDescent

Definition at line 55 of file itkFiniteDifferenceGradientDescentOptimizer.h.

Inheritance diagram for itk::FiniteDifferenceGradientDescentOptimizer:

[legend]

Public Types
using	ConstPointer = SmartPointer<const Self>

using	Pointer = SmartPointer<Self>

using	Self = FiniteDifferenceGradientDescentOptimizer

enum	StopConditionType { MaximumNumberOfIterations , MetricError }

using	Superclass = ScaledSingleValuedNonLinearOptimizer

Public Types inherited from itk::ScaledSingleValuedNonLinearOptimizer
using	ConstPointer = SmartPointer<const Self>

using	Pointer = SmartPointer<Self>

using	ScaledCostFunctionPointer = ScaledCostFunctionType::Pointer

using	ScaledCostFunctionType = ScaledSingleValuedCostFunction

using	ScalesType = NonLinearOptimizer::ScalesType

using	Self = ScaledSingleValuedNonLinearOptimizer

using	Superclass = SingleValuedNonLinearOptimizer

Public Member Functions
virtual void	AdvanceOneStep ()

virtual void	ComputeCurrentValueOff ()

virtual void	ComputeCurrentValueOn ()

virtual const char *	GetClassName () const

virtual bool	GetComputeCurrentValue () const

virtual unsigned long	GetCurrentIteration () const

virtual double	GetGradientMagnitude () const

virtual double	GetLearningRate () const

virtual unsigned long	GetNumberOfIterations () const

virtual double	GetParam_A () const

virtual double	GetParam_a () const

virtual double	GetParam_alpha () const

virtual double	GetParam_c () const

virtual double	GetParam_gamma () const

virtual StopConditionType	GetStopCondition () const

virtual double	GetValue () const

	ITK_DISALLOW_COPY_AND_MOVE (FiniteDifferenceGradientDescentOptimizer)

void	ResumeOptimization ()

virtual void	SetComputeCurrentValue (bool _arg)

virtual void	SetNumberOfIterations (unsigned long _arg)

virtual void	SetParam_A (double _arg)

virtual void	SetParam_a (double _arg)

virtual void	SetParam_alpha (double _arg)

virtual void	SetParam_c (double _arg)

virtual void	SetParam_gamma (double _arg)

void	StartOptimization () override

void	StopOptimization ()

Public Member Functions inherited from itk::ScaledSingleValuedNonLinearOptimizer
const ParametersType &	GetCurrentPosition () const override

virtual bool	GetMaximize () const

virtual const ScaledCostFunctionType *	GetScaledCostFunction ()

virtual const ParametersType &	GetScaledCurrentPosition ()

bool	GetUseScales () const

virtual void	InitializeScales ()

	ITK_DISALLOW_COPY_AND_MOVE (ScaledSingleValuedNonLinearOptimizer)

virtual void	MaximizeOff ()

virtual void	MaximizeOn ()

void	SetCostFunction (CostFunctionType *costFunction) override

virtual void	SetMaximize (bool _arg)

virtual void	SetUseScales (bool arg)

Static Public Member Functions
static Pointer	New ()

Static Public Member Functions inherited from itk::ScaledSingleValuedNonLinearOptimizer
static Pointer	New ()

Protected Member Functions
virtual double	Compute_a (unsigned long k) const

virtual double	Compute_c (unsigned long k) const

	FiniteDifferenceGradientDescentOptimizer ()

void	PrintSelf (std::ostream &os, Indent indent) const override

	~FiniteDifferenceGradientDescentOptimizer () override=default

Protected Member Functions inherited from itk::ScaledSingleValuedNonLinearOptimizer
virtual void	GetScaledDerivative (const ParametersType &parameters, DerivativeType &derivative) const

virtual MeasureType	GetScaledValue (const ParametersType &parameters) const

virtual void	GetScaledValueAndDerivative (const ParametersType &parameters, MeasureType &value, DerivativeType &derivative) const

void	PrintSelf (std::ostream &os, Indent indent) const override

	ScaledSingleValuedNonLinearOptimizer ()

void	SetCurrentPosition (const ParametersType &param) override

virtual void	SetScaledCurrentPosition (const ParametersType &parameters)

	~ScaledSingleValuedNonLinearOptimizer () override=default

Protected Attributes
bool	m_ComputeCurrentValue { false }

DerivativeType	m_Gradient {}

double	m_GradientMagnitude { 0.0 }

double	m_LearningRate { 0.0 }

Protected Attributes inherited from itk::ScaledSingleValuedNonLinearOptimizer
ScaledCostFunctionPointer	m_ScaledCostFunction {}

ParametersType	m_ScaledCurrentPosition {}

Private Attributes
unsigned long	m_CurrentIteration { 0 }

unsigned long	m_NumberOfIterations { 100 }

double	m_Param_A { 1.0 }

double	m_Param_a { 1.0 }

double	m_Param_alpha { 0.602 }

double	m_Param_c { 1.0 }

double	m_Param_gamma { 0.101 }

bool	m_Stop { false }

StopConditionType	m_StopCondition { MaximumNumberOfIterations }

double	m_Value { 0.0 }

Member Typedef Documentation

◆ ConstPointer

using itk::FiniteDifferenceGradientDescentOptimizer::ConstPointer = SmartPointer<const Self>

Definition at line 64 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ Pointer

using itk::FiniteDifferenceGradientDescentOptimizer::Pointer = SmartPointer<Self>

Definition at line 63 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ Self

using itk::FiniteDifferenceGradientDescentOptimizer::Self = FiniteDifferenceGradientDescentOptimizer

Standard class typedefs.

Definition at line 61 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ Superclass

using itk::FiniteDifferenceGradientDescentOptimizer::Superclass = ScaledSingleValuedNonLinearOptimizer

Definition at line 62 of file itkFiniteDifferenceGradientDescentOptimizer.h.

Member Enumeration Documentation

◆ StopConditionType

enum itk::FiniteDifferenceGradientDescentOptimizer::StopConditionType

Codes of stopping conditions

Enumerator
MaximumNumberOfIterations
MetricError

Definition at line 73 of file itkFiniteDifferenceGradientDescentOptimizer.h.

Constructor & Destructor Documentation

◆ FiniteDifferenceGradientDescentOptimizer()

itk::FiniteDifferenceGradientDescentOptimizer::FiniteDifferenceGradientDescentOptimizer ( )

protected

◆ ~FiniteDifferenceGradientDescentOptimizer()

itk::FiniteDifferenceGradientDescentOptimizer::~FiniteDifferenceGradientDescentOptimizer ( )

overrideprotecteddefault

Member Function Documentation

◆ AdvanceOneStep()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::AdvanceOneStep ( )

virtual

Advance one step following the gradient direction.

◆ Compute_a()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::Compute_a ( unsigned long k ) const

protectedvirtual

◆ Compute_c()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::Compute_c ( unsigned long k ) const

protectedvirtual

◆ ComputeCurrentValueOff()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::ComputeCurrentValueOff ( )

virtual

◆ ComputeCurrentValueOn()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::ComputeCurrentValueOn ( )

virtual

◆ GetClassName()

virtual const char * itk::FiniteDifferenceGradientDescentOptimizer::GetClassName ( ) const

virtual

Run-time type information (and related methods).

Reimplemented from itk::ScaledSingleValuedNonLinearOptimizer.

Reimplemented in elastix::FiniteDifferenceGradientDescent< TElastix >.

◆ GetComputeCurrentValue()

virtual bool itk::FiniteDifferenceGradientDescentOptimizer::GetComputeCurrentValue ( ) const

virtual

◆ GetCurrentIteration()

virtual unsigned long itk::FiniteDifferenceGradientDescentOptimizer::GetCurrentIteration ( ) const

virtual

Get the current iteration number.

◆ GetGradientMagnitude()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetGradientMagnitude ( ) const

virtual

Get the CurrentStepLength, GradientMagnitude and LearningRate (a_k)

◆ GetLearningRate()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetLearningRate ( ) const

virtual

◆ GetNumberOfIterations()

virtual unsigned long itk::FiniteDifferenceGradientDescentOptimizer::GetNumberOfIterations ( ) const

virtual

Get the number of iterations.

◆ GetParam_A()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetParam_A ( ) const

virtual

◆ GetParam_a()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetParam_a ( ) const

virtual

◆ GetParam_alpha()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetParam_alpha ( ) const

virtual

◆ GetParam_c()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetParam_c ( ) const

virtual

◆ GetParam_gamma()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetParam_gamma ( ) const

virtual

◆ GetStopCondition()

virtual StopConditionType itk::FiniteDifferenceGradientDescentOptimizer::GetStopCondition ( ) const

virtual

Get Stop condition.

◆ GetValue()

virtual double itk::FiniteDifferenceGradientDescentOptimizer::GetValue ( ) const

virtual

Get the current value.

◆ ITK_DISALLOW_COPY_AND_MOVE()

itk::FiniteDifferenceGradientDescentOptimizer::ITK_DISALLOW_COPY_AND_MOVE ( FiniteDifferenceGradientDescentOptimizer )

◆ New()

static Pointer itk::FiniteDifferenceGradientDescentOptimizer::New ( )

static

Method for creation through the object factory.

◆ PrintSelf()

void itk::FiniteDifferenceGradientDescentOptimizer::PrintSelf	(	std::ostream &	os,
		Indent	indent ) const

overrideprotected

PrintSelf method.

◆ ResumeOptimization()

void itk::FiniteDifferenceGradientDescentOptimizer::ResumeOptimization ( )

Resume previously stopped optimization with current parameters

See also: StopOptimization.

◆ SetComputeCurrentValue()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetComputeCurrentValue ( bool _arg )

virtual

◆ SetNumberOfIterations()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetNumberOfIterations ( unsigned long _arg )

virtual

Set the number of iterations.

◆ SetParam_A()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetParam_A ( double _arg )

virtual

Set/Get A.

◆ SetParam_a()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetParam_a ( double _arg )

virtual

Set/Get a.

◆ SetParam_alpha()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetParam_alpha ( double _arg )

virtual

Set/Get alpha.

◆ SetParam_c()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetParam_c ( double _arg )

virtual

Set/Get c.

◆ SetParam_gamma()

virtual void itk::FiniteDifferenceGradientDescentOptimizer::SetParam_gamma ( double _arg )

virtual

Set/Get gamma.

◆ StartOptimization()

void itk::FiniteDifferenceGradientDescentOptimizer::StartOptimization ( )

override

Start optimization.

◆ StopOptimization()

void itk::FiniteDifferenceGradientDescentOptimizer::StopOptimization ( )

Stop optimization.

See also: ResumeOptimization

Field Documentation

◆ m_ComputeCurrentValue

bool itk::FiniteDifferenceGradientDescentOptimizer::m_ComputeCurrentValue { false }

protected

Boolean that says if the current value of the metric has to be computed. This is not necessary for optimisation; just nice for progress information.

Definition at line 158 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_CurrentIteration

unsigned long itk::FiniteDifferenceGradientDescentOptimizer::m_CurrentIteration { 0 }

private

Definition at line 173 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Gradient

DerivativeType itk::FiniteDifferenceGradientDescentOptimizer::m_Gradient {}

protected

Definition at line 149 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_GradientMagnitude

double itk::FiniteDifferenceGradientDescentOptimizer::m_GradientMagnitude { 0.0 }

protected

Definition at line 151 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_LearningRate

double itk::FiniteDifferenceGradientDescentOptimizer::m_LearningRate { 0.0 }

protected

Definition at line 150 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_NumberOfIterations

unsigned long itk::FiniteDifferenceGradientDescentOptimizer::m_NumberOfIterations { 100 }

private

Definition at line 172 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Param_A

double itk::FiniteDifferenceGradientDescentOptimizer::m_Param_A { 1.0 }

private

Definition at line 178 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Param_a

double itk::FiniteDifferenceGradientDescentOptimizer::m_Param_a { 1.0 }

private

Parameters, as described by Spall.

Definition at line 176 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Param_alpha

double itk::FiniteDifferenceGradientDescentOptimizer::m_Param_alpha { 0.602 }

private

Definition at line 179 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Param_c

double itk::FiniteDifferenceGradientDescentOptimizer::m_Param_c { 1.0 }

private

Definition at line 177 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Param_gamma

double itk::FiniteDifferenceGradientDescentOptimizer::m_Param_gamma { 0.101 }

private

Definition at line 180 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Stop

bool itk::FiniteDifferenceGradientDescentOptimizer::m_Stop { false }

private

Private member variables.

Definition at line 169 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_StopCondition

StopConditionType itk::FiniteDifferenceGradientDescentOptimizer::m_StopCondition { MaximumNumberOfIterations }

private

Definition at line 171 of file itkFiniteDifferenceGradientDescentOptimizer.h.

◆ m_Value

double itk::FiniteDifferenceGradientDescentOptimizer::m_Value { 0.0 }

private

Definition at line 170 of file itkFiniteDifferenceGradientDescentOptimizer.h.

Detailed Description

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Protected Attributes

Private Attributes

Member Typedef Documentation

◆ ConstPointer

◆ Pointer

◆ Self

◆ Superclass

Member Enumeration Documentation

◆ StopConditionType

Constructor & Destructor Documentation

◆ FiniteDifferenceGradientDescentOptimizer()

◆ ~FiniteDifferenceGradientDescentOptimizer()

Member Function Documentation

◆ AdvanceOneStep()

◆ Compute_a()

◆ Compute_c()

◆ ComputeCurrentValueOff()

◆ ComputeCurrentValueOn()

◆ GetClassName()

◆ GetComputeCurrentValue()

◆ GetCurrentIteration()

◆ GetGradientMagnitude()

◆ GetLearningRate()

◆ GetNumberOfIterations()

◆ GetParam_A()

◆ GetParam_a()

◆ GetParam_alpha()

◆ GetParam_c()

◆ GetParam_gamma()

◆ GetStopCondition()

◆ GetValue()

◆ ITK_DISALLOW_COPY_AND_MOVE()

◆ New()

◆ PrintSelf()

◆ ResumeOptimization()

◆ SetComputeCurrentValue()

◆ SetNumberOfIterations()

◆ SetParam_A()

◆ SetParam_a()

◆ SetParam_alpha()

◆ SetParam_c()

◆ SetParam_gamma()

◆ StartOptimization()

◆ StopOptimization()

Field Documentation

◆ m_ComputeCurrentValue

◆ m_CurrentIteration

◆ m_Gradient

◆ m_GradientMagnitude

◆ m_LearningRate

◆ m_NumberOfIterations

◆ m_Param_A

◆ m_Param_a

◆ m_Param_alpha

◆ m_Param_c

◆ m_Param_gamma

◆ m_Stop

◆ m_StopCondition

◆ m_Value