This class offers methods for the usage of the Resilient-Backpropagation-algorithm without weight-backtracking. More...

#include <shark/Algorithms/GradientDescent/Rprop.h>

Inheritance diagram for shark::RpropMinus:

Public Member Functions
SHARK_EXPORT_SYMBOL	RpropMinus ()

std::string	name () const
	From INameable: return the class name. More...

SHARK_EXPORT_SYMBOL void	init (ObjectiveFunctionType const &objectiveFunction, SearchPointType const &startingPoint)
	initializes the optimizer using a predefined starting point More...

virtual SHARK_EXPORT_SYMBOL void	init (ObjectiveFunctionType const &objectiveFunction, SearchPointType const &startingPoint, double initDelta)

SHARK_EXPORT_SYMBOL void	step (ObjectiveFunctionType const &objectiveFunction)
	Carry out one step of the optimizer for the supplied objective function. More...

virtual SHARK_EXPORT_SYMBOL void	read (InArchive &archive)
	Read the component from the supplied archive. More...

virtual SHARK_EXPORT_SYMBOL void	write (OutArchive &archive) const
	Write the component to the supplied archive. More...

void	setEtaMinus (double etaMinus)
	set decrease factor More...

void	setEtaPlus (double etaPlus)
	set increase factor More...

void	setMaxDelta (double d)
	set upper limit on update More...

void	setMinDelta (double d)
	set lower limit on update More...

double	maxDelta () const
	return the maximal step size component More...

RealVector const &	derivative () const
	Returns the derivative at the current point. Can be used for stopping criteria. More...

Public Member Functions inherited from shark::AbstractSingleObjectiveOptimizer< RealVector >
std::size_t	numInitPoints () const
	By default most single objective optimizers only require a single point. More...

virtual void	init (ObjectiveFunctionType const &function, std::vector< SearchPointType > const &initPoints)
	Initialize the optimizer for the supplied objective function using a set of initialisation points. More...

virtual const SolutionType &	solution () const
	returns the current solution of the optimizer More...

Public Member Functions inherited from shark::AbstractOptimizer< RealVector, double, SingleObjectiveResultSet< RealVector > >
const Features &	features () const

virtual void	updateFeatures ()

bool	requiresValue () const

bool	requiresFirstDerivative () const

bool	requiresSecondDerivative () const

bool	canSolveConstrained () const

bool	requiresClosestFeasible () const

virtual	~AbstractOptimizer ()

virtual void	init (ObjectiveFunctionType const &function)
	Initialize the optimizer for the supplied objective function. More...

Public Member Functions inherited from shark::INameable
virtual	~INameable ()

Public Member Functions inherited from shark::ISerializable
virtual	~ISerializable ()
	Virtual d'tor. More...

void	load (InArchive &archive, unsigned int version)
	Versioned loading of components, calls read(...). More...

void	save (OutArchive &archive, unsigned int version) const
	Versioned storing of components, calls write(...). More...

	BOOST_SERIALIZATION_SPLIT_MEMBER ()

Protected Attributes
ObjectiveFunctionType::FirstOrderDerivative	m_derivative

double	m_increaseFactor
	The increase factor \( \eta^+ \), set to 1.2 by default. More...

double	m_decreaseFactor
	The decrease factor \( \eta^- \), set to 0.5 by default. More...

double	m_maxDelta
	The upper limit of the increments \( \Delta w_i^{(t)} \), set to 1e100 by default. More...

double	m_minDelta
	The lower limit of the increments \( \Delta w_i^{(t)} \), set to 0.0 by default. More...

size_t	m_parameterSize

RealVector	m_oldDerivative
	The last error gradient. More...

RealVector	m_delta
	The absolute update values (increment) for all weights. More...

Protected Attributes inherited from shark::AbstractSingleObjectiveOptimizer< RealVector >
SolutionType	m_best
	Current solution of the optimizer. More...

Protected Attributes inherited from shark::AbstractOptimizer< RealVector, double, SingleObjectiveResultSet< RealVector > >
Features	m_features

Additional Inherited Members
Public Types inherited from shark::AbstractSingleObjectiveOptimizer< RealVector >
typedef base_type::SearchPointType	SearchPointType

typedef base_type::SolutionType	SolutionType

typedef base_type::ResultType	ResultType

typedef base_type::ObjectiveFunctionType	ObjectiveFunctionType

Public Types inherited from shark::AbstractOptimizer< RealVector, double, SingleObjectiveResultSet< RealVector > >
enum	Feature
	Models features that the optimizer requires from the objective function. More...

typedef RealVector	SearchPointType

typedef double	ResultType

typedef SingleObjectiveResultSet< RealVector >	SolutionType

typedef AbstractObjectiveFunction< RealVector, ResultType >	ObjectiveFunctionType

typedef TypedFlags< Feature >	Features

typedef TypedFeatureNotAvailableException< Feature >	FeatureNotAvailableException

Protected Member Functions inherited from shark::AbstractOptimizer< RealVector, double, SingleObjectiveResultSet< RealVector > >
void	checkFeatures (ObjectiveFunctionType const &objectiveFunction)
	Convenience function that checks whether the features of the supplied objective function match with the required features of the optimizer. More...

Detailed Description

This class offers methods for the usage of the Resilient-Backpropagation-algorithm without weight-backtracking.

The Rprop algorithm is an improvement of the algorithms with adaptive learning rates (as the Adaptive Backpropagation algorithm by Silva and Ameida, please see AdpBP.h for a description of the working of such an algorithm), that uses increments for the update of the weights, that are independent from the absolute partial derivatives. This makes sense, because large flat regions in the search space (plateaus) lead to small absolute partial derivatives and so the increments are chosen small, but the increments should be large to skip the plateau. In contrast, the absolute partial derivatives are very large at the "slopes" of very "narrow canyons", which leads to large increments that will skip the minimum lying at the bottom of the canyon, but it would make more sense to chose small increments to hit the minimum.
So, the Rprop algorithm only uses the signs of the partial derivatives and not the absolute values to adapt the parameters.
Instead of individual learning rates, it uses the parameter \(\Delta_i^{(t)}\) for weight \(w_i,\ i = 1, \dots, n\) in iteration \(t\), where the parameter will be adapted before the change of the weights:

\( \Delta_i^{(t)} = \Bigg\{ \begin{array}{ll} min( \eta^+ \cdot \Delta_i^{(t-1)}, \Delta_{max} ), & \mbox{if\ } \frac{\partial E^{(t-1)}}{\partial w_i} \cdot \frac{\partial E^{(t)}}{\partial w_i} > 0 \\ max( \eta^- \cdot \Delta_i^{(t-1)}, \Delta_{min} ), & \mbox{if\ } \frac{\partial E^{(t-1)}}{\partial w_i} \cdot \frac{\partial E^{(t)}}{\partial w_i} < 0 \\ \Delta_i^{(t-1)}, & \mbox{otherwise} \end{array} \)

The parameters \(\eta^+ > 1\) and \(0 < \eta^- < 1\) control the speed of the adaptation. To stabilize the increments, they are restricted to the interval \([\Delta_{min}, \Delta_{max}]\).
After the adaptation of the \(\Delta_i\) the update for the weights will be calculated as

\( \Delta w_i^{(t)} := - \mbox{sign} \left( \frac{\partial E^{(t)}}{\partial w_i}\right) \cdot \Delta_i^{(t)} \)

For further information about the algorithm, please refer to:

Martin Riedmiller,
"Advanced Supervised Learning in Multi-layer Perceptrons - From Backpropagation to Adaptive Learning Algorithms".
In "International Journal of Computer Standards and Interfaces", volume 16, no. 5, 1994, pp. 265-278

Author: C. Igel

Date: 1999

Changes:: none

Status:: stable

Definition at line 108 of file Rprop.h.

Constructor & Destructor Documentation

◆ RpropMinus()

SHARK_EXPORT_SYMBOL shark::RpropMinus::RpropMinus ( )

Member Function Documentation

◆ derivative()

RealVector const& shark::RpropMinus::derivative ( ) const

inline

Returns the derivative at the current point. Can be used for stopping criteria.

Definition at line 157 of file Rprop.h.

References m_derivative.

◆ init() [1/2]

SHARK_EXPORT_SYMBOL void shark::RpropMinus::init	(	ObjectiveFunctionType const &	function,
		SearchPointType const &	startingPoint
	)

virtual

initializes the optimizer using a predefined starting point

Implements shark::AbstractSingleObjectiveOptimizer< RealVector >.

Reimplemented in shark::IRpropPlusFull, shark::IRpropPlus, and shark::RpropPlus.

Referenced by name(), shark::RpropPlus::name(), shark::IRpropPlus::name(), and shark::IRpropPlusFull::name().

◆ init() [2/2]

virtual SHARK_EXPORT_SYMBOL void shark::RpropMinus::init	(	ObjectiveFunctionType const &	objectiveFunction,
		SearchPointType const &	startingPoint,
		double	initDelta
	)

virtual

Reimplemented in shark::IRpropPlusFull, shark::IRpropPlus, and shark::RpropPlus.

◆ maxDelta()

double shark::RpropMinus::maxDelta ( ) const

inline

return the maximal step size component

Definition at line 152 of file Rprop.h.

References m_delta.

Referenced by run_one_trial().

◆ name()

std::string shark::RpropMinus::name ( ) const

inlinevirtual

From INameable: return the class name.

Reimplemented from shark::INameable.

Reimplemented in shark::IRpropMinus, shark::IRpropPlusFull, shark::IRpropPlus, and shark::RpropPlus.

Definition at line 114 of file Rprop.h.

References init(), read(), SHARK_EXPORT_SYMBOL, step(), and write().

◆ read()

virtual SHARK_EXPORT_SYMBOL void shark::RpropMinus::read ( InArchive & archive )

virtual

Read the component from the supplied archive.

Parameters

[in,out] archive The archive to read from.

Reimplemented from shark::ISerializable.

Reimplemented in shark::IRpropPlusFull, shark::IRpropPlus, and shark::RpropPlus.

Referenced by name(), shark::RpropPlus::name(), shark::IRpropPlus::name(), and shark::IRpropPlusFull::name().

◆ setEtaMinus()

void shark::RpropMinus::setEtaMinus ( double etaMinus )

inline

set decrease factor

Definition at line 127 of file Rprop.h.

References m_decreaseFactor, and RANGE_CHECK.

◆ setEtaPlus()

void shark::RpropMinus::setEtaPlus ( double etaPlus )

inline

set increase factor

Definition at line 134 of file Rprop.h.

References m_increaseFactor, and RANGE_CHECK.

◆ setMaxDelta()

void shark::RpropMinus::setMaxDelta ( double d )

inline

set upper limit on update

Definition at line 140 of file Rprop.h.

References m_maxDelta, and RANGE_CHECK.

◆ setMinDelta()

void shark::RpropMinus::setMinDelta ( double d )

inline

set lower limit on update

Definition at line 146 of file Rprop.h.

References m_minDelta, and RANGE_CHECK.

◆ step()

SHARK_EXPORT_SYMBOL void shark::RpropMinus::step ( ObjectiveFunctionType const & function )

virtual

Carry out one step of the optimizer for the supplied objective function.

Parameters

[in] function The objective function to initialize for.

Implements shark::AbstractOptimizer< RealVector, double, SingleObjectiveResultSet< RealVector > >.

Reimplemented in shark::IRpropMinus, shark::IRpropPlusFull, shark::IRpropPlus, and shark::RpropPlus.

Referenced by name(), shark::RpropPlus::name(), shark::IRpropPlus::name(), shark::IRpropPlusFull::name(), and shark::IRpropMinus::name().

◆ write()

virtual SHARK_EXPORT_SYMBOL void shark::RpropMinus::write ( OutArchive & archive ) const

virtual

Write the component to the supplied archive.

Parameters

[in,out] archive The archive to write to.

Reimplemented from shark::ISerializable.

Reimplemented in shark::IRpropPlusFull, shark::IRpropPlus, and shark::RpropPlus.

Referenced by name(), shark::RpropPlus::name(), shark::IRpropPlus::name(), and shark::IRpropPlusFull::name().

Member Data Documentation

◆ m_decreaseFactor

double shark::RpropMinus::m_decreaseFactor

protected

The decrease factor \( \eta^- \), set to 0.5 by default.

Definition at line 167 of file Rprop.h.

Referenced by setEtaMinus().

◆ m_delta

RealVector shark::RpropMinus::m_delta

protected

The absolute update values (increment) for all weights.

Definition at line 181 of file Rprop.h.

Referenced by maxDelta().

◆ m_derivative

ObjectiveFunctionType::FirstOrderDerivative shark::RpropMinus::m_derivative

protected

Definition at line 161 of file Rprop.h.

Referenced by derivative().

◆ m_increaseFactor

double shark::RpropMinus::m_increaseFactor

protected

The increase factor \( \eta^+ \), set to 1.2 by default.

Definition at line 164 of file Rprop.h.

Referenced by setEtaPlus().

◆ m_maxDelta

double shark::RpropMinus::m_maxDelta

protected

The upper limit of the increments \( \Delta w_i^{(t)} \), set to 1e100 by default.

Definition at line 170 of file Rprop.h.

Referenced by setMaxDelta().

◆ m_minDelta

double shark::RpropMinus::m_minDelta

protected

The lower limit of the increments \( \Delta w_i^{(t)} \), set to 0.0 by default.

Definition at line 173 of file Rprop.h.

Referenced by setMinDelta().

◆ m_oldDerivative

RealVector shark::RpropMinus::m_oldDerivative

protected

The last error gradient.

Definition at line 178 of file Rprop.h.

◆ m_parameterSize

size_t shark::RpropMinus::m_parameterSize

protected

Definition at line 175 of file Rprop.h.

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

include/shark/Algorithms/GradientDescent/Rprop.h

Public Member Functions

Protected Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ RpropMinus()

Member Function Documentation

◆ derivative()

◆ init() [1/2]

◆ init() [2/2]

◆ maxDelta()

◆ name()

◆ read()

◆ setEtaMinus()

◆ setEtaPlus()

◆ setMaxDelta()

◆ setMinDelta()

◆ step()

◆ write()

Member Data Documentation

◆ m_decreaseFactor

◆ m_delta

◆ m_derivative

◆ m_increaseFactor

◆ m_maxDelta

◆ m_minDelta

◆ m_oldDerivative

◆ m_parameterSize