itk::SPSAOptimizer Class Reference

#include <itkSPSAOptimizer.h>

Inheritance diagram for itk::SPSAOptimizer:

Collaboration diagram for itk::SPSAOptimizer:

Detailed Description

An optimizer based on simultaneous perturbation...

This optimizer is an implementation of the Simultaneous Perturbation Stochastic Approximation method, described in:

• http://www.jhuapl.edu/SPSA/
• Spall, J.C. (1998), "An Overview of the Simultaneous Perturbation Method for Efficient Optimization," Johns Hopkins APL Technical Digest, vol. 19, pp. 482-492

Definition at line 44 of file itkSPSAOptimizer.h.

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef SmartPointer< Self >	Pointer
typedef SPSAOptimizer	Self
enum	StopConditionType {   Unknown,   MaximumNumberOfIterations,   BelowTolerance,   MetricError }
typedef SingleValuedNonLinearOptimizer	Superclass
Public Types inherited from itk::SingleValuedNonLinearOptimizer
typedef SmartPointer< const Self >	ConstPointer
typedef CostFunctionType::Pointer	CostFunctionPointer
typedef SingleValuedCostFunction	CostFunctionType
typedef CostFunctionType::DerivativeType	DerivativeType
typedef CostFunctionType::MeasureType	MeasureType
typedef Superclass::ParametersType	ParametersType
typedef SmartPointer< Self >	Pointer
typedef SingleValuedNonLinearOptimizer	Self
typedef NonLinearOptimizer	Superclass
Public Types inherited from itk::NonLinearOptimizer
typedef SmartPointer< const Self >	ConstPointer
typedef Superclass::ParametersType	ParametersType
typedef SmartPointer< Self >	Pointer
typedef Superclass::ScalesType	ScalesType
typedef NonLinearOptimizer	Self
typedef Optimizer	Superclass
Public Types inherited from itk::Optimizer
typedef SmartPointer< const Self >	ConstPointer
typedef OptimizerParameters < double >	ParametersType
typedef SmartPointer< Self >	Pointer
typedef Array< double >	ScalesType
typedef Optimizer	Self
typedef Object	Superclass
Public Types inherited from itk::Object
typedef SmartPointer< const Self >	ConstPointer
typedef SmartPointer< Self >	Pointer
typedef Object	Self
typedef LightObject	Superclass
Public Types inherited from itk::LightObject
typedef SmartPointer< const Self >	ConstPointer
typedef SmartPointer< Self >	Pointer
typedef LightObject	Self

Public Member Functions
virtual void	AdvanceOneStep ()
virtual ::itk::LightObject::Pointer	CreateAnother () const
virtual SizeValueType	GetCurrentIteration () const
virtual const DerivativeType &	GetGradient () const
virtual double	GetGradientMagnitude () const
virtual double	GetLearningRate () const
virtual const char *	GetNameOfClass () const
virtual double	GetStateOfConvergence () const
virtual StopConditionType	GetStopCondition () const
virtual const std::string	GetStopConditionDescription () const override
virtual MeasureType	GetValue () const
virtual MeasureType	GetValue (const ParametersType &parameters) const
virtual void	GuessParameters (SizeValueType numberOfGradientEstimates, double initialStepSize)
void	ResumeOptimization ()
virtual void	StartOptimization (void) override
void	StopOptimization ()
virtual void	SetSa (double _arg)
virtual double	GetSa () const
void	Seta (double a)
double	Geta ()
virtual void	SetSc (double _arg)
virtual double	GetSc () const
void	Setc (double c)
double	Getc ()
virtual void	SetA (double _arg)
virtual double	GetA () const
virtual void	SetAlpha (double _arg)
virtual double	GetAlpha () const
virtual void	SetGamma (double _arg)
virtual double	GetGamma () const
virtual bool	GetMaximize () const
virtual void	SetMaximize (bool _arg)
virtual void	MaximizeOn ()
virtual void	MaximizeOff ()
bool	GetMinimize () const
void	SetMinimize (bool v)
void	MinimizeOn ()
void	MinimizeOff ()
virtual void	SetNumberOfPerturbations (SizeValueType _arg)
virtual SizeValueType	GetNumberOfPerturbations () const
virtual void	SetStateOfConvergenceDecayRate (double _arg)
virtual double	GetStateOfConvergenceDecayRate () const
virtual void	SetMinimumNumberOfIterations (SizeValueType _arg)
virtual SizeValueType	GetMinimumNumberOfIterations () const
virtual void	SetMaximumNumberOfIterations (SizeValueType _arg)
virtual SizeValueType	GetMaximumNumberOfIterations () const
virtual void	SetTolerance (double _arg)
virtual double	GetTolerance () const
Public Member Functions inherited from itk::SingleValuedNonLinearOptimizer
virtual const CostFunctionType *	GetCostFunction () const
virtual CostFunctionType *	GetModifiableCostFunction ()
MeasureType	GetValue (const ParametersType &parameters) const
virtual void	SetCostFunction (CostFunctionType *costFunction)
Public Member Functions inherited from itk::Optimizer
virtual const ParametersType &	GetCurrentPosition () const
virtual const ParametersType &	GetInitialPosition () const
virtual void	SetInitialPosition (const ParametersType &param)
void	SetScales (const ScalesType &scales)
virtual const ScalesType &	GetScales () const
virtual const ScalesType &	GetInverseScales () const
Public Member Functions inherited from itk::Object
unsigned long	AddObserver (const EventObject &event, Command *)
unsigned long	AddObserver (const EventObject &event, Command *) const
virtual void	DebugOff () const
virtual void	DebugOn () const
Command *	GetCommand (unsigned long tag)
bool	GetDebug () const
MetaDataDictionary &	GetMetaDataDictionary ()
const MetaDataDictionary &	GetMetaDataDictionary () const
virtual ModifiedTimeType	GetMTime () const
virtual const TimeStamp &	GetTimeStamp () const
bool	HasObserver (const EventObject &event) const
void	InvokeEvent (const EventObject &)
void	InvokeEvent (const EventObject &) const
virtual void	Modified () const
virtual void	Register () const override
void	RemoveAllObservers ()
void	RemoveObserver (unsigned long tag)
void	SetDebug (bool debugFlag) const
void	SetMetaDataDictionary (const MetaDataDictionary &rhs)
virtual void	SetReferenceCount (int) override
virtual void	UnRegister () const noexceptoverride
virtual void	SetObjectName (std::string _arg)
virtual const std::string &	GetObjectName () const
Public Member Functions inherited from itk::LightObject
virtual void	Delete ()
virtual int	GetReferenceCount () const
	itkCloneMacro (Self)
void	Print (std::ostream &os, Indent indent=0) const

Static Public Member Functions
static Pointer	New ()
Static Public Member Functions inherited from itk::SingleValuedNonLinearOptimizer
static Pointer	New ()
Static Public Member Functions inherited from itk::NonLinearOptimizer
static Pointer	New ()
Static Public Member Functions inherited from itk::Optimizer
static Pointer	New ()
Static Public Member Functions inherited from itk::Object
static bool	GetGlobalWarningDisplay ()
static void	GlobalWarningDisplayOff ()
static void	GlobalWarningDisplayOn ()
static Pointer	New ()
static void	SetGlobalWarningDisplay (bool flag)
Static Public Member Functions inherited from itk::LightObject
static void	BreakOnError ()
static Pointer	New ()

Protected Member Functions
virtual double	Compute_a (SizeValueType k) const
virtual double	Compute_c (SizeValueType k) const
virtual void	ComputeGradient (const ParametersType &parameters, DerivativeType &gradient)
virtual void	GenerateDelta (const unsigned int spaceDimension)
virtual void	PrintSelf (std::ostream &os, Indent indent) const override
	SPSAOptimizer ()
virtual	~SPSAOptimizer () override
Protected Member Functions inherited from itk::SingleValuedNonLinearOptimizer
	SingleValuedNonLinearOptimizer ()
virtual	~SingleValuedNonLinearOptimizer () override
Protected Member Functions inherited from itk::NonLinearOptimizer
	NonLinearOptimizer ()
virtual	~NonLinearOptimizer () override
Protected Member Functions inherited from itk::Optimizer
	Optimizer ()
virtual void	SetCurrentPosition (const ParametersType &param)
virtual	~Optimizer () override
Protected Member Functions inherited from itk::Object
	Object ()
bool	PrintObservers (std::ostream &os, Indent indent) const
virtual void	SetTimeStamp (const TimeStamp &time)
virtual	~Object () override
Protected Member Functions inherited from itk::LightObject
virtual LightObject::Pointer	InternalClone () const
	LightObject ()
virtual void	PrintHeader (std::ostream &os, Indent indent) const
virtual void	PrintTrailer (std::ostream &os, Indent indent) const
virtual	~LightObject ()

Protected Attributes
SizeValueType	m_CurrentIteration
DerivativeType	m_Delta
Statistics::MersenneTwisterRandomVariateGenerator::Pointer	m_Generator
DerivativeType	m_Gradient
double	m_LearningRate
double	m_StateOfConvergence
bool	m_Stop
StopConditionType	m_StopCondition
Protected Attributes inherited from itk::SingleValuedNonLinearOptimizer
CostFunctionPointer	m_CostFunction
Protected Attributes inherited from itk::Optimizer
ParametersType	m_CurrentPosition
bool	m_ScalesInitialized
Protected Attributes inherited from itk::LightObject
AtomicInt< int >	m_ReferenceCount

Private Attributes
double	m_A
double	m_Alpha
double	m_Gamma
double	m_GradientMagnitude
bool	m_Maximize
SizeValueType	m_MaximumNumberOfIterations
SizeValueType	m_MinimumNumberOfIterations
SizeValueType	m_NumberOfPerturbations
double	m_Sa
double	m_Sc
double	m_StateOfConvergenceDecayRate
double	m_Tolerance

Member Typedef Documentation

typedef SmartPointer< const Self > itk::SPSAOptimizer::ConstPointer

Definition at line 53 of file itkSPSAOptimizer.h.

typedef SmartPointer< Self > itk::SPSAOptimizer::Pointer

Definition at line 52 of file itkSPSAOptimizer.h.

typedef SPSAOptimizer itk::SPSAOptimizer::Self

Standard class typedefs.

Definition at line 50 of file itkSPSAOptimizer.h.

typedef SingleValuedNonLinearOptimizer itk::SPSAOptimizer::Superclass

Definition at line 51 of file itkSPSAOptimizer.h.

Member Enumeration Documentation

enum itk::SPSAOptimizer::StopConditionType

Codes of stopping conditions

Enumerator
Unknown
MaximumNumberOfIterations
BelowTolerance
MetricError

Definition at line 62 of file itkSPSAOptimizer.h.

Constructor & Destructor Documentation

itk::SPSAOptimizer::SPSAOptimizer ( )

protected

virtual itk::SPSAOptimizer::~SPSAOptimizer ( )

inlineoverrideprotectedvirtual

Definition at line 214 of file itkSPSAOptimizer.h.

Member Function Documentation

virtual void itk::SPSAOptimizer::AdvanceOneStep ( )

virtual

Advance one step following the gradient direction.

virtual double itk::SPSAOptimizer::Compute_a ( SizeValueType  k ) const

protectedvirtual

Method to compute the learning rate at iteration k (a_k).

virtual double itk::SPSAOptimizer::Compute_c ( SizeValueType  k ) const

protectedvirtual

Method to compute the gain factor for the perturbation at iteration k (c_k).

virtual void itk::SPSAOptimizer::ComputeGradient ( const ParametersType &  parameters, DerivativeType &  gradient  )

protectedvirtual

Compute the gradient at a position. m_NumberOfPerturbations are used, and scales are taken into account.

virtual::itk::LightObject::Pointer itk::SPSAOptimizer::CreateAnother ( ) const

virtual

Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

virtual void itk::SPSAOptimizer::GenerateDelta ( const unsigned int  spaceDimension )

protectedvirtual

Method to generate a perturbation vector. Takes scales into account.

double itk::SPSAOptimizer::Geta ( )

inline

Set/Get a.

Definition at line 126 of file itkSPSAOptimizer.h.

virtual double itk::SPSAOptimizer::GetA ( ) const

virtual

Set/Get A.

virtual double itk::SPSAOptimizer::GetAlpha ( ) const

virtual

Set/Get alpha.

double itk::SPSAOptimizer::Getc ( )

inline

Set/Get c.

Definition at line 134 of file itkSPSAOptimizer.h.

virtual SizeValueType itk::SPSAOptimizer::GetCurrentIteration ( ) const

virtual

Get the current iteration number.

virtual double itk::SPSAOptimizer::GetGamma ( ) const

virtual

Set/Get gamma.

virtual const DerivativeType& itk::SPSAOptimizer::GetGradient ( ) const

virtual

Get the latest computed gradient

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

virtual

Get the GradientMagnitude of the latest computed gradient

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

virtual

Get the current LearningRate (a_k)

virtual bool itk::SPSAOptimizer::GetMaximize ( ) const

virtual

Methods to configure the cost function.

virtual SizeValueType itk::SPSAOptimizer::GetMaximumNumberOfIterations ( ) const

virtual

Set/Get the maximum number of iterations.

bool itk::SPSAOptimizer::GetMinimize ( ) const

inline

Methods to configure the cost function.

Definition at line 156 of file itkSPSAOptimizer.h.

virtual SizeValueType itk::SPSAOptimizer::GetMinimumNumberOfIterations ( ) const

virtual

Set/Get the minimum number of iterations

virtual const char* itk::SPSAOptimizer::GetNameOfClass ( ) const

virtual

Run-time type information (and related methods).

Reimplemented from itk::SingleValuedNonLinearOptimizer.

virtual SizeValueType itk::SPSAOptimizer::GetNumberOfPerturbations ( ) const

virtual

Set/Get the number of perturbation used to construct a gradient estimate g_k. q = NumberOfPerturbations g_k = 1/q sum_{j=1..q} g^(j)_k

virtual double itk::SPSAOptimizer::GetSa ( ) const

virtual

Set/Get a.

virtual double itk::SPSAOptimizer::GetSc ( ) const

virtual

Set/Get c.

virtual double itk::SPSAOptimizer::GetStateOfConvergence ( ) const

virtual

Get the state of convergence in the last iteration. When the StateOfConvergence is lower than the Tolerance, and the minimum number of iterations has been performed, the optimization stops.

The state of convergence (SOC) is initialized with 0.0 and updated after each iteration as follows: SOC *= SOCDecayRate SOC += a_k * GradientMagnitude

virtual double itk::SPSAOptimizer::GetStateOfConvergenceDecayRate ( ) const

virtual

Set/Get StateOfConvergenceDecayRate (number between 0 and 1).

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

virtual

Get Stop condition.

virtual const std::string itk::SPSAOptimizer::GetStopConditionDescription ( ) const

overridevirtual

Get the reason for termination

Reimplemented from itk::Optimizer.

virtual double itk::SPSAOptimizer::GetTolerance ( ) const

virtual

Set/Get Tolerance

virtual MeasureType itk::SPSAOptimizer::GetValue ( ) const

virtual

Get the cost function value at the current position.

virtual MeasureType itk::SPSAOptimizer::GetValue ( const ParametersType &  parameters ) const

virtual

Get the cost function value at any position

virtual void itk::SPSAOptimizer::GuessParameters ( SizeValueType  numberOfGradientEstimates, double  initialStepSize  )

virtual

Guess the parameters a and A. This function needs the number of GradientEstimates used for estimating a and A and and the expected initial step size (where step size is defined as the maximum of the absolute values of the parameter update). Make sure you set c, Alpha, Gamma, the MaximumNumberOfIterations, the Scales, and the the InitialPosition before calling this method.

Described in: Spall, J.C. (1998), "Implementation of the Simultaneous Perturbation Algorithm for Stochastic Optimization", IEEE Trans. Aerosp. Electron. Syst. 34(3), 817-823.

virtual void itk::SPSAOptimizer::MaximizeOff ( )

virtual

Methods to configure the cost function.

virtual void itk::SPSAOptimizer::MaximizeOn ( )

virtual

Methods to configure the cost function.

void itk::SPSAOptimizer::MinimizeOff ( )

inline

Methods to configure the cost function.

Definition at line 162 of file itkSPSAOptimizer.h.

void itk::SPSAOptimizer::MinimizeOn ( )

inline

Methods to configure the cost function.

Definition at line 160 of file itkSPSAOptimizer.h.

static Pointer itk::SPSAOptimizer::New ( )

static

Method for creation through the object factory.

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

overrideprotectedvirtual

PrintSelf method.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

void itk::SPSAOptimizer::ResumeOptimization

(

)

Resume previously stopped optimization with current parameters

See Also

StopOptimization.

void itk::SPSAOptimizer::Seta ( double  a )

inline

Set/Get a.

Definition at line 125 of file itkSPSAOptimizer.h.

virtual void itk::SPSAOptimizer::SetA ( double  _arg )

virtual

Set/Get A.

virtual void itk::SPSAOptimizer::SetAlpha ( double  _arg )

virtual

Set/Get alpha.

void itk::SPSAOptimizer::Setc ( double  c )

inline

Set/Get c.

Definition at line 133 of file itkSPSAOptimizer.h.

virtual void itk::SPSAOptimizer::SetGamma ( double  _arg )

virtual

Set/Get gamma.

virtual void itk::SPSAOptimizer::SetMaximize ( bool  _arg )

virtual

Methods to configure the cost function.

virtual void itk::SPSAOptimizer::SetMaximumNumberOfIterations ( SizeValueType  _arg )

virtual

Set/Get the maximum number of iterations.

void itk::SPSAOptimizer::SetMinimize ( bool  v )

inline

Methods to configure the cost function.

Definition at line 158 of file itkSPSAOptimizer.h.

virtual void itk::SPSAOptimizer::SetMinimumNumberOfIterations ( SizeValueType  _arg )

virtual

Set/Get the minimum number of iterations

virtual void itk::SPSAOptimizer::SetNumberOfPerturbations ( SizeValueType  _arg )

virtual

Set/Get the number of perturbation used to construct a gradient estimate g_k. q = NumberOfPerturbations g_k = 1/q sum_{j=1..q} g^(j)_k

virtual void itk::SPSAOptimizer::SetSa ( double  _arg )

virtual

Set/Get a.

virtual void itk::SPSAOptimizer::SetSc ( double  _arg )

virtual

Set/Get c.

virtual void itk::SPSAOptimizer::SetStateOfConvergenceDecayRate ( double  _arg )

virtual

Set/Get StateOfConvergenceDecayRate (number between 0 and 1).

virtual void itk::SPSAOptimizer::SetTolerance ( double  _arg )

virtual

Set/Get Tolerance

virtual void itk::SPSAOptimizer::StartOptimization ( void  )

overridevirtual

Start optimization.

Reimplemented from itk::Optimizer.

void itk::SPSAOptimizer::StopOptimization

(

)

Stop optimization.

See Also

ResumeOptimization

Member Data Documentation

double itk::SPSAOptimizer::m_A

private

Definition at line 273 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_Alpha

private

Definition at line 274 of file itkSPSAOptimizer.h.

SizeValueType itk::SPSAOptimizer::m_CurrentIteration

protected

Definition at line 232 of file itkSPSAOptimizer.h.

DerivativeType itk::SPSAOptimizer::m_Delta

protected

Definition at line 224 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_Gamma

private

Definition at line 275 of file itkSPSAOptimizer.h.

Statistics::MersenneTwisterRandomVariateGenerator::Pointer itk::SPSAOptimizer::m_Generator

protected

Random number generator

Definition at line 235 of file itkSPSAOptimizer.h.

DerivativeType itk::SPSAOptimizer::m_Gradient

protected

Variables updated during optimization

Definition at line 220 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_GradientMagnitude

private

Definition at line 267 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_LearningRate

protected

Definition at line 222 of file itkSPSAOptimizer.h.

bool itk::SPSAOptimizer::m_Maximize

private

Definition at line 266 of file itkSPSAOptimizer.h.

SizeValueType itk::SPSAOptimizer::m_MaximumNumberOfIterations

private

Definition at line 263 of file itkSPSAOptimizer.h.

SizeValueType itk::SPSAOptimizer::m_MinimumNumberOfIterations

private

Settings.

Definition at line 259 of file itkSPSAOptimizer.h.

SizeValueType itk::SPSAOptimizer::m_NumberOfPerturbations

private

Definition at line 268 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_Sa

private

Parameters, as described by Spall.

Definition at line 271 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_Sc

private

Definition at line 272 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_StateOfConvergence

protected

Definition at line 230 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_StateOfConvergenceDecayRate

private

Definition at line 264 of file itkSPSAOptimizer.h.

bool itk::SPSAOptimizer::m_Stop

protected

Definition at line 226 of file itkSPSAOptimizer.h.

StopConditionType itk::SPSAOptimizer::m_StopCondition

protected

Definition at line 228 of file itkSPSAOptimizer.h.

double itk::SPSAOptimizer::m_Tolerance

private

Definition at line 265 of file itkSPSAOptimizer.h.

---

The documentation for this class was generated from the following file:

• itkSPSAOptimizer.h