shark::CMA Class Reference

Implements the CMA-ES. More...

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

+ Inheritance diagram for shark::CMA:

Public Types

enum  RecombinationType { EQUAL = 0, LINEAR = 1, SUPERLINEAR = 2 }
 Models the recombination type. More...
 
- 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, ResultTypeObjectiveFunctionType
 
typedef TypedFlags< FeatureFeatures
 
typedef TypedFeatureNotAvailableException< FeatureFeatureNotAvailableException
 

Public Member Functions

SHARK_EXPORT_SYMBOL CMA (DefaultRngType &rng=Rng::globalRng)
 Default c'tor. More...
 
std::string name () const
 From INameable: return the class name. More...
 
template<typename Container , typename Extractor >
RealVector weightedSum (const Container &container, const RealVector &weights, const Extractor &e)
 Calculates the center of gravity of the given population \( \in \mathbb{R}^d\). More...
 
SHARK_EXPORT_SYMBOL void read (InArchive &archive)
 Read the component from the supplied archive. More...
 
SHARK_EXPORT_SYMBOL void write (OutArchive &archive) const
 Write the component to the supplied archive. More...
 
SHARK_EXPORT_SYMBOL void init (ObjectiveFunctionType &function, SearchPointType const &p)
 Initializes the algorithm for the supplied objective function. More...
 
SHARK_EXPORT_SYMBOL void init (ObjectiveFunctionType &function, SearchPointType const &initialSearchPoint, std::size_t lambda, std::size_t mu, double initialSigma, const boost::optional< RealMatrix > &initialCovarianceMatrix=boost::optional< RealMatrix >())
 Initializes the algorithm for the supplied objective function. More...
 
SHARK_EXPORT_SYMBOL void step (ObjectiveFunctionType const &function)
 Executes one iteration of the algorithm. More...
 
void setInitialSigma (double initSigma)
 sets the initial step length sigma More...
 
double sigma () const
 Accesses the current step size. More...
 
RealVector const & mean () const
 Accesses the current population mean. More...
 
RealVector const & weights () const
 Accesses the current weighting vector. More...
 
RealVector const & evolutionPath () const
 Accesses the evolution path for the covariance matrix update. More...
 
RealVector const & evolutionPathSigma () const
 Accesses the evolution path for the step size update. More...
 
RealMatrix const & covarianceMatrix () const
 Accesses the covariance matrix of the normal distribution used for generating offspring individuals. More...
 
RecombinationType recombinationType () const
 Accesses the recombination type. More...
 
RecombinationTyperecombinationType ()
 Returns a mutable reference to the recombination type. More...
 
const double & lowerBound () const
 Returns a const reference tothe lower bound on sigma times smalles eigenvalue. More...
 
std::size_t mu () const
 Returns the size of the parent population \(\mu\). More...
 
void setMu (std::size_t mu)
 Sets the number of selected samples. More...
 
void setLambda (std::size_t lambda)
 Sets the number of sampled points. More...
 
std::size_t lambda () const
 Returns a immutable reference to the size of the offspring population \(\mu\). More...
 
RealMatrix const & eigenVectors () const
 Returns eigenvectors of covariance matrix (not considering step size) More...
 
RealVector const & eigenValues () const
 Returns a eigenvectors of covariance matrix (not considering step size) More...
 
double condition () const
 Returns condition of covariance matrix. More...
 
- Public Member Functions inherited from shark::AbstractSingleObjectiveOptimizer< RealVector >
virtual void init (ObjectiveFunctionType &function)
 
virtual const SolutionTypesolution () const
 returns the current solution of the optimizer More...
 
- Public Member Functions inherited from shark::AbstractOptimizer< RealVector, double, SingleObjectiveResultSet< RealVector > >
const Featuresfeatures () const
 
virtual void updateFeatures ()
 
bool requiresValue () const
 
bool requiresFirstDerivative () const
 
bool requiresSecondDerivative () const
 
bool canSolveConstrained () const
 
bool requiresClosestFeasible () const
 
virtual ~AbstractOptimizer ()
 Empty virtual d'tor. 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 ()
 

Static Public Member Functions

static SHARK_EXPORT_SYMBOL std::size_t suggestLambda (std::size_t dimension)
 Calculates lambda for the supplied dimensionality n. More...
 
static SHARK_EXPORT_SYMBOL std::size_t suggestMu (std::size_t lambda, RecombinationType recomb=SUPERLINEAR)
 Calculates mu for the supplied lambda and the recombination strategy. More...
 

Protected Types

typedef Individual< RealVector, double, RealVector > IndividualType
 The type of individual used for the CMA. More...
 

Protected Member Functions

SHARK_EXPORT_SYMBOL std::vector< IndividualTypegenerateOffspring () const
 Samples lambda individuals from the search distribution. More...
 
SHARK_EXPORT_SYMBOL void updatePopulation (std::vector< IndividualType > const &offspring)
 Updates the strategy parameters based on the supplied offspring population. More...
 
SHARK_EXPORT_SYMBOL void doInit (std::vector< SearchPointType > const &points, std::vector< ResultType > const &functionValues, std::size_t lambda, std::size_t mu, double initialSigma)
 
- 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...
 

Additional Inherited Members

- 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
 

Detailed Description

Implements the CMA-ES.

The algorithm is described in

Hansen, N., S. Kern (2004). Evaluating the CMA Evolution Strategy on Multimodal Test Functions. In Proceedings of the Eighth International Conference on Parallel Problem Solving from Nature (PPSN VIII), pp. 282-291, LNCS, Springer-Verlag

Definition at line 57 of file CMA.h.

Member Typedef Documentation

§ IndividualType

typedef Individual<RealVector, double, RealVector> shark::CMA::IndividualType
protected

The type of individual used for the CMA.

Definition at line 238 of file CMA.h.

Member Enumeration Documentation

§ RecombinationType

Models the recombination type.

Enumerator
EQUAL 
LINEAR 
SUPERLINEAR 

Definition at line 63 of file CMA.h.

Constructor & Destructor Documentation

§ CMA()

SHARK_EXPORT_SYMBOL shark::CMA::CMA ( DefaultRngType rng = Rng::globalRng)

Default c'tor.

Member Function Documentation

§ condition()

double shark::CMA::condition ( ) const
inline

Returns condition of covariance matrix.

Definition at line 230 of file CMA.h.

References shark::MultiVariateNormalDistribution::eigenValues(), eigenValues(), shark::blas::max(), and shark::blas::min().

Referenced by main().

§ covarianceMatrix()

RealMatrix const& shark::CMA::covarianceMatrix ( ) const
inline

Accesses the covariance matrix of the normal distribution used for generating offspring individuals.

Definition at line 163 of file CMA.h.

References shark::MultiVariateNormalDistribution::covarianceMatrix().

§ doInit()

SHARK_EXPORT_SYMBOL void shark::CMA::doInit ( std::vector< SearchPointType > const &  points,
std::vector< ResultType > const &  functionValues,
std::size_t  lambda,
std::size_t  mu,
double  initialSigma 
)
protected

§ eigenValues()

RealVector const& shark::CMA::eigenValues ( ) const
inline

Returns a eigenvectors of covariance matrix (not considering step size)

Definition at line 223 of file CMA.h.

References shark::MultiVariateNormalDistribution::eigenValues().

Referenced by condition(), and main().

§ eigenVectors()

RealMatrix const& shark::CMA::eigenVectors ( ) const
inline

Returns eigenvectors of covariance matrix (not considering step size)

Definition at line 216 of file CMA.h.

References shark::MultiVariateNormalDistribution::eigenVectors().

Referenced by main().

§ evolutionPath()

RealVector const& shark::CMA::evolutionPath ( ) const
inline

Accesses the evolution path for the covariance matrix update.

Definition at line 153 of file CMA.h.

§ evolutionPathSigma()

RealVector const& shark::CMA::evolutionPathSigma ( ) const
inline

Accesses the evolution path for the step size update.

Definition at line 158 of file CMA.h.

§ generateOffspring()

SHARK_EXPORT_SYMBOL std::vector<IndividualType> shark::CMA::generateOffspring ( ) const
protected

Samples lambda individuals from the search distribution.

§ init() [1/2]

SHARK_EXPORT_SYMBOL void shark::CMA::init ( ObjectiveFunctionType function,
SearchPointType const &  p 
)
virtual

Initializes the algorithm for the supplied objective function.

Implements shark::AbstractSingleObjectiveOptimizer< RealVector >.

Referenced by main(), and weightedSum().

§ init() [2/2]

SHARK_EXPORT_SYMBOL void shark::CMA::init ( ObjectiveFunctionType function,
SearchPointType const &  initialSearchPoint,
std::size_t  lambda,
std::size_t  mu,
double  initialSigma,
const boost::optional< RealMatrix > &  initialCovarianceMatrix = boost::optional< RealMatrix >() 
)

Initializes the algorithm for the supplied objective function.

§ lambda()

std::size_t shark::CMA::lambda ( ) const
inline

Returns a immutable reference to the size of the offspring population \(\mu\).

Definition at line 209 of file CMA.h.

Referenced by setLambda(), and weightedSum().

§ lowerBound()

const double& shark::CMA::lowerBound ( ) const
inline

Returns a const reference tothe lower bound on sigma times smalles eigenvalue.

Definition at line 184 of file CMA.h.

§ mean()

RealVector const& shark::CMA::mean ( ) const
inline

Accesses the current population mean.

Definition at line 143 of file CMA.h.

Referenced by main().

§ mu()

std::size_t shark::CMA::mu ( ) const
inline

Returns the size of the parent population \(\mu\).

Definition at line 191 of file CMA.h.

Referenced by setMu(), and weightedSum().

§ name()

std::string shark::CMA::name ( ) const
inlinevirtual

From INameable: return the class name.

Reimplemented from shark::INameable.

Definition at line 75 of file CMA.h.

§ read()

SHARK_EXPORT_SYMBOL void shark::CMA::read ( InArchive archive)
virtual

Read the component from the supplied archive.

Parameters
[in,out]archiveThe archive to read from.

Reimplemented from shark::ISerializable.

Referenced by weightedSum().

§ recombinationType() [1/2]

RecombinationType shark::CMA::recombinationType ( ) const
inline

Accesses the recombination type.

Definition at line 170 of file CMA.h.

§ recombinationType() [2/2]

RecombinationType& shark::CMA::recombinationType ( )
inline

Returns a mutable reference to the recombination type.

Definition at line 177 of file CMA.h.

§ setInitialSigma()

void shark::CMA::setInitialSigma ( double  initSigma)
inline

sets the initial step length sigma

It is by default <=0 which means that sigma =1/sqrt(numVariables)

Definition at line 133 of file CMA.h.

Referenced by main().

§ setLambda()

void shark::CMA::setLambda ( std::size_t  lambda)
inline

Sets the number of sampled points.

Definition at line 201 of file CMA.h.

References lambda().

§ setMu()

void shark::CMA::setMu ( std::size_t  mu)
inline

Sets the number of selected samples.

Definition at line 196 of file CMA.h.

References mu().

§ sigma()

double shark::CMA::sigma ( ) const
inline

Accesses the current step size.

Definition at line 138 of file CMA.h.

Referenced by main().

§ step()

SHARK_EXPORT_SYMBOL void shark::CMA::step ( ObjectiveFunctionType const &  function)
virtual

Executes one iteration of the algorithm.

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

Referenced by main(), and weightedSum().

§ suggestLambda()

static SHARK_EXPORT_SYMBOL std::size_t shark::CMA::suggestLambda ( std::size_t  dimension)
static

Calculates lambda for the supplied dimensionality n.

Referenced by weightedSum().

§ suggestMu()

static SHARK_EXPORT_SYMBOL std::size_t shark::CMA::suggestMu ( std::size_t  lambda,
RecombinationType  recomb = SUPERLINEAR 
)
static

Calculates mu for the supplied lambda and the recombination strategy.

Referenced by weightedSum().

§ updatePopulation()

SHARK_EXPORT_SYMBOL void shark::CMA::updatePopulation ( std::vector< IndividualType > const &  offspring)
protected

Updates the strategy parameters based on the supplied offspring population.

§ weightedSum()

template<typename Container , typename Extractor >
RealVector shark::CMA::weightedSum ( const Container &  container,
const RealVector &  weights,
const Extractor &  e 
)
inline

Calculates the center of gravity of the given population \( \in \mathbb{R}^d\).

Definition at line 84 of file CMA.h.

References init(), lambda(), mu(), read(), SHARK_EXPORT_SYMBOL, step(), suggestLambda(), suggestMu(), SUPERLINEAR, weights(), and write().

§ weights()

RealVector const& shark::CMA::weights ( ) const
inline

Accesses the current weighting vector.

Definition at line 148 of file CMA.h.

Referenced by weightedSum().

§ write()

SHARK_EXPORT_SYMBOL void shark::CMA::write ( OutArchive archive) const
virtual

Write the component to the supplied archive.

Parameters
[in,out]archiveThe archive to write to.

Reimplemented from shark::ISerializable.

Referenced by weightedSum().


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