Implements the \((\mu+1)\)-MO-CMA-ES. More...

#include <shark/Algorithms/DirectSearch/SteadyStateMOCMA.h>

Inheritance diagram for shark::IndicatorBasedSteadyStateMOCMA< Indicator >:

Public Types
enum	NotionOfSuccess { IndividualBased, PopulationBased }

Public Types inherited from shark::AbstractMultiObjectiveOptimizer< RealVector >
typedef super::SearchPointType	SearchPointType

typedef super::SolutionType	SolutionType

typedef super::ObjectiveFunctionType	ObjectiveFunctionType

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

typedef RealVector	SearchPointType

typedef RealVector	ResultType

typedef std::vector< ResultSet< RealVector, RealVector > >	SolutionType

typedef AbstractObjectiveFunction< RealVector, ResultType >	ObjectiveFunctionType

typedef TypedFlags< Feature >	Features

typedef TypedFeatureNotAvailableException< Feature >	FeatureNotAvailableException

Public Member Functions
	IndicatorBasedSteadyStateMOCMA (random::rng_type &rng=random::globalRng)

std::string	name () const
	Returns the name of the algorithm. More...

std::size_t	mu () const

std::size_t &	mu ()

std::size_t	numInitPoints () const
	Returns the number of points this method requires for initialisation. More...

double	initialSigma () const

double &	initialSigma ()

NotionOfSuccess	notionOfSuccess () const

NotionOfSuccess &	notionOfSuccess ()

Indicator &	indicator ()

Indicator const &	indicator () const

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

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

void	init (ObjectiveFunctionType const &function, std::vector< SearchPointType > const &initialSearchPoints)
	Initializes the algorithm for the supplied objective function. More...

void	step (ObjectiveFunctionType const &function)
	Executes one iteration of the algorithm. More...

Public Member Functions inherited from shark::AbstractMultiObjectiveOptimizer< RealVector >
SolutionType const &	solution () const
	Accesses the current approximation of the Pareto-set and -front, respectively. More...

Public Member Functions inherited from shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, 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 Types
typedef CMAIndividual< RealVector >	IndividualType
	The individual type of the SteadyState-MOCMA. More...

Protected Member Functions
void	doInit (std::vector< SearchPointType > const &initialSearchPoints, std::vector< ResultType > const &functionValues, std::size_t mu, double initialSigma)

std::vector< IndividualType >	generateOffspring () const

void	updatePopulation (std::vector< IndividualType > const &offspringVec)

Protected Member Functions inherited from shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, 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...

Protected Attributes
std::vector< IndividualType >	m_parents
	Population of size \(\mu + 1\). More...

Protected Attributes inherited from shark::AbstractMultiObjectiveOptimizer< RealVector >
SolutionType	m_best
	The current Pareto-set/-front. More...

Protected Attributes inherited from shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, RealVector > > >
Features	m_features

Detailed Description

template<typename Indicator = HypervolumeIndicator>
class shark::IndicatorBasedSteadyStateMOCMA< Indicator >

Implements the \((\mu+1)\)-MO-CMA-ES.

Please see the following papers for further reference:

Igel, Suttorp and Hansen. Steady-state Selection and Efficient Covariance Matrix Update in the Multi-Objective CMA-ES.
Vo�, Hansen and Igel. Improved Step Size Adaptation for the MO-CMA-ES.

Definition at line 52 of file SteadyStateMOCMA.h.

Member Typedef Documentation

◆ IndividualType

template<typename Indicator = HypervolumeIndicator>

typedef CMAIndividual<RealVector> shark::IndicatorBasedSteadyStateMOCMA< Indicator >::IndividualType

protected

The individual type of the SteadyState-MOCMA.

Definition at line 161 of file SteadyStateMOCMA.h.

Member Enumeration Documentation

◆ NotionOfSuccess

template<typename Indicator = HypervolumeIndicator>

enum shark::IndicatorBasedSteadyStateMOCMA::NotionOfSuccess

Enumerator
IndividualBased
PopulationBased

Definition at line 54 of file SteadyStateMOCMA.h.

Constructor & Destructor Documentation

◆ IndicatorBasedSteadyStateMOCMA()

template<typename Indicator = HypervolumeIndicator>

shark::IndicatorBasedSteadyStateMOCMA< Indicator >::IndicatorBasedSteadyStateMOCMA ( random::rng_type & rng = random::globalRng )

inline

Definition at line 60 of file SteadyStateMOCMA.h.

References shark::IndicatorBasedSteadyStateMOCMA< Indicator >::initialSigma(), shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, RealVector > > >::m_features, shark::IndicatorBasedSteadyStateMOCMA< Indicator >::mu(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::notionOfSuccess(), and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::PopulationBased.

Member Function Documentation

◆ doInit()

template<typename Indicator = HypervolumeIndicator>

void shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit	(	std::vector< SearchPointType > const &	initialSearchPoints,
		std::vector< ResultType > const &	functionValues,
		std::size_t	mu,
		double	initialSigma
	)

inlineprotected

Definition at line 163 of file SteadyStateMOCMA.h.

References shark::random::discrete(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::indicator(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::initialSigma(), shark::AbstractMultiObjectiveOptimizer< RealVector >::m_best, shark::IndicatorBasedSteadyStateMOCMA< Indicator >::m_parents, shark::IndicatorBasedSteadyStateMOCMA< Indicator >::mu(), and SIZE_CHECK.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::init().

◆ generateOffspring()

template<typename Indicator = HypervolumeIndicator>

std::vector<IndividualType> shark::IndicatorBasedSteadyStateMOCMA< Indicator >::generateOffspring ( ) const

inlineprotected

Definition at line 207 of file SteadyStateMOCMA.h.

References shark::random::discrete(), and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::m_parents.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::step().

◆ indicator() [1/2]

template<typename Indicator = HypervolumeIndicator>

Indicator& shark::IndicatorBasedSteadyStateMOCMA< Indicator >::indicator ( )

inline

Definition at line 100 of file SteadyStateMOCMA.h.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit().

◆ indicator() [2/2]

template<typename Indicator = HypervolumeIndicator>

Indicator const& shark::IndicatorBasedSteadyStateMOCMA< Indicator >::indicator ( ) const

inline

Definition at line 103 of file SteadyStateMOCMA.h.

◆ init()

template<typename Indicator = HypervolumeIndicator>

void shark::IndicatorBasedSteadyStateMOCMA< Indicator >::init	(	ObjectiveFunctionType const &	function,
		std::vector< SearchPointType > const &	initialSearchPoints
	)

inlinevirtual

Initializes the algorithm for the supplied objective function.

Parameters

[in]	function	The objective function.
[in]	initialSearchPoints	A set of intiial search points.

Implements shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, RealVector > > >.

Definition at line 135 of file SteadyStateMOCMA.h.

References shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, RealVector > > >::checkFeatures(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::initialSigma(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::mu(), and SHARK_RUNTIME_CHECK.

◆ initialSigma() [1/2]

template<typename Indicator = HypervolumeIndicator>

double shark::IndicatorBasedSteadyStateMOCMA< Indicator >::initialSigma ( ) const

inline

Definition at line 86 of file SteadyStateMOCMA.h.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::IndicatorBasedSteadyStateMOCMA(), and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::init().

◆ initialSigma() [2/2]

template<typename Indicator = HypervolumeIndicator>

double& shark::IndicatorBasedSteadyStateMOCMA< Indicator >::initialSigma ( )

inline

Definition at line 89 of file SteadyStateMOCMA.h.

◆ mu() [1/2]

template<typename Indicator = HypervolumeIndicator>

std::size_t shark::IndicatorBasedSteadyStateMOCMA< Indicator >::mu ( ) const

inline

Definition at line 75 of file SteadyStateMOCMA.h.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::IndicatorBasedSteadyStateMOCMA(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::init(), and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::updatePopulation().

◆ mu() [2/2]

template<typename Indicator = HypervolumeIndicator>

std::size_t& shark::IndicatorBasedSteadyStateMOCMA< Indicator >::mu ( )

inline

Definition at line 78 of file SteadyStateMOCMA.h.

◆ name()

template<typename Indicator = HypervolumeIndicator>

std::string shark::IndicatorBasedSteadyStateMOCMA< Indicator >::name ( ) const

inlinevirtual

Returns the name of the algorithm.

Reimplemented from shark::INameable.

Definition at line 71 of file SteadyStateMOCMA.h.

◆ notionOfSuccess() [1/2]

template<typename Indicator = HypervolumeIndicator>

NotionOfSuccess shark::IndicatorBasedSteadyStateMOCMA< Indicator >::notionOfSuccess ( ) const

inline

Definition at line 93 of file SteadyStateMOCMA.h.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::IndicatorBasedSteadyStateMOCMA().

◆ notionOfSuccess() [2/2]

template<typename Indicator = HypervolumeIndicator>

NotionOfSuccess& shark::IndicatorBasedSteadyStateMOCMA< Indicator >::notionOfSuccess ( )

inline

Definition at line 96 of file SteadyStateMOCMA.h.

◆ numInitPoints()

template<typename Indicator = HypervolumeIndicator>

std::size_t shark::IndicatorBasedSteadyStateMOCMA< Indicator >::numInitPoints ( ) const

inlinevirtual

Returns the number of points this method requires for initialisation.

The number of points supplied is allowed to be smaller, however in this case the optimizer will resort to techniques generating additional points if needed.

Implements shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, RealVector > > >.

Definition at line 82 of file SteadyStateMOCMA.h.

◆ read()

template<typename Indicator = HypervolumeIndicator>

void shark::IndicatorBasedSteadyStateMOCMA< Indicator >::read ( InArchive & archive )

inlinevirtual

Read the component from the supplied archive.

Parameters

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

Reimplemented from shark::ISerializable.

Definition at line 107 of file SteadyStateMOCMA.h.

References shark::AbstractMultiObjectiveOptimizer< RealVector >::m_best, and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::m_parents.

◆ step()

template<typename Indicator = HypervolumeIndicator>

void shark::IndicatorBasedSteadyStateMOCMA< Indicator >::step ( ObjectiveFunctionType const & function )

inlinevirtual

Executes one iteration of the algorithm.

Parameters

[in] function The function to iterate upon.

Implements shark::AbstractOptimizer< RealVector, RealVector, std::vector< ResultSet< RealVector, RealVector > > >.

Definition at line 153 of file SteadyStateMOCMA.h.

References shark::IndicatorBasedSteadyStateMOCMA< Indicator >::generateOffspring(), and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::updatePopulation().

◆ updatePopulation()

template<typename Indicator = HypervolumeIndicator>

void shark::IndicatorBasedSteadyStateMOCMA< Indicator >::updatePopulation ( std::vector< IndividualType > const & offspringVec )

inlineprotected

Definition at line 223 of file SteadyStateMOCMA.h.

References shark::IndicatorBasedSteadyStateMOCMA< Indicator >::IndividualBased, shark::AbstractMultiObjectiveOptimizer< RealVector >::m_best, shark::IndicatorBasedSteadyStateMOCMA< Indicator >::m_parents, shark::IndicatorBasedSteadyStateMOCMA< Indicator >::mu(), shark::CMAIndividual< FitnessType >::parent(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::PopulationBased, shark::Individual< PointType, FitnessTypeT, Chromosome >::rank(), shark::Individual< PointType, FitnessTypeT, Chromosome >::selected(), shark::CMAChromosome::Successful, shark::CMAChromosome::Unsuccessful, shark::CMAIndividual< FitnessType >::updateAsOffspring(), and shark::CMAIndividual< FitnessType >::updateAsParent().

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::step().

◆ write()

template<typename Indicator = HypervolumeIndicator>

void shark::IndicatorBasedSteadyStateMOCMA< Indicator >::write ( OutArchive & archive ) const

inlinevirtual

Write the component to the supplied archive.

Parameters

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

Reimplemented from shark::ISerializable.

Definition at line 117 of file SteadyStateMOCMA.h.

References shark::AbstractMultiObjectiveOptimizer< RealVector >::m_best, and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::m_parents.

Member Data Documentation

◆ m_parents

template<typename Indicator = HypervolumeIndicator>

std::vector<IndividualType> shark::IndicatorBasedSteadyStateMOCMA< Indicator >::m_parents

protected

Population of size \(\mu + 1\).

Definition at line 256 of file SteadyStateMOCMA.h.

Referenced by shark::IndicatorBasedSteadyStateMOCMA< Indicator >::doInit(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::generateOffspring(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::read(), shark::IndicatorBasedSteadyStateMOCMA< Indicator >::updatePopulation(), and shark::IndicatorBasedSteadyStateMOCMA< Indicator >::write().

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

include/shark/Algorithms/DirectSearch/SteadyStateMOCMA.h

Public Types

Public Member Functions

Protected Types

Protected Member Functions

Protected Attributes

Detailed Description

template<typename Indicator = HypervolumeIndicator> class shark::IndicatorBasedSteadyStateMOCMA< Indicator >

Member Typedef Documentation

◆ IndividualType

Member Enumeration Documentation

◆ NotionOfSuccess

Constructor & Destructor Documentation

◆ IndicatorBasedSteadyStateMOCMA()

Member Function Documentation

◆ doInit()

◆ generateOffspring()

◆ indicator() [1/2]

◆ indicator() [2/2]

◆ init()

◆ initialSigma() [1/2]

◆ initialSigma() [2/2]

◆ mu() [1/2]

◆ mu() [2/2]

◆ name()

◆ notionOfSuccess() [1/2]

◆ notionOfSuccess() [2/2]

◆ numInitPoints()

◆ read()

◆ step()

◆ updatePopulation()

◆ write()

Member Data Documentation

◆ m_parents

template<typename Indicator = HypervolumeIndicator>
class shark::IndicatorBasedSteadyStateMOCMA< Indicator >