Particle Swarm optimization generational with racing. More...

#include <pso_generational_racing.h>

Inheritance diagram for pagmo::algorithm::pso_generational_racing:

Public Member Functions
	pso_generational_racing (int gen=1, double omega=0.7298, double eta1=2.05, double eta2=2.05, double vcoeff=0.5, int variant=5, int neighb_type=2, int neighb_param=4, unsigned int nr_eval_per_x=5, unsigned int max_fevals=std::numeric_limits< unsigned int >::max())
	Constructor. More...

base_ptr	clone () const
	Clone method.

void	evolve (population &) const
	Evolve implementation. More...

std::string	get_name () const
	Algorithm name.

Public Member Functions inherited from pagmo::algorithm::base
	base ()
	Default constructor. More...

virtual	~base ()
	Trivial destructor. More...

std::string	human_readable () const
	Return human readable representation of the algorithm. More...

void	set_screen_output (const bool p)
	Setter-Getter for protected m_screen_output data. More...

bool	get_screen_output () const
	Gets screen output. More...

void	reset_rngs (const unsigned int) const
	Resets the seed of the internal rngs using a user-provided seed. More...

Protected Member Functions
std::string	human_readable_extra () const
	Extra human readable algorithm info. More...

Friends
class	boost::serialization::access

Additional Inherited Members
Protected Attributes inherited from pagmo::algorithm::base
bool	m_screen_output
	Indicates to the derived class whether to print stuff on screen.

rng_double	m_drng
	Random number generator for double-precision floating point values.

rng_uint32	m_urng
	Random number generator for unsigned integer values.

unsigned int	m_fevals
	A counter for the number of function evaluations.

Detailed Description

Particle Swarm optimization generational with racing.

Compared to the generational version of Particle Swarm Optimization, this PSO is further extended with racing mechanisms for individuals.

Definition at line 43 of file pso_generational_racing.h.

Constructor & Destructor Documentation

pagmo::algorithm::pso_generational_racing::pso_generational_racing	(	int	gen = `1`,
		double	omega = `0.7298`,
		double	eta1 = `2.05`,
		double	eta2 = `2.05`,
		double	vcoeff = `0.5`,
		int	variant = `5`,
		int	neighb_type = `2`,
		int	neighb_param = `4`,
		unsigned int	nr_eval_per_x = `5`,
		unsigned int	max_fevals = `std::numeric_limits<unsigned int>::max()`
	)

Constructor.

Allows to specify in detail all the parameters of the algorithm.

Parameters

[in]	gen	number of generations
[in]	omega	particles' inertia weight, or alternatively, the constriction coefficient (usage depends on the variant used)
[in]	eta1	magnitude of the force, applied to the particle's velocity, in the direction of its previous best position
[in]	eta2	magnitude of the force, applied to the particle's velocity, in the direction of the best position in its neighborhood
[in]	vcoeff	velocity coefficient (determining the maximum allowed particle velocity)
[in]	variant	algorithm variant to use
[in]	neighb_type	swarm topology to use
[in]	neighb_param	if the lbest topology is selected (neighb_type=2), it represents each particle's indegree (also outdegree) in the swarm topology. Particles have neighbours u to a radius of k = neighb_param / 2 in the ring. If the Randomly-varying neighbourhood topology is selected (neighb_type=4), neighb_param represents each particle's maximum outdegree in the swarm topology. The minimum outdegree is 1 (the particle always connects back to itself).
[in]	nr_eval_per_x	Expected number of times an objective function will be evaluated for each individual during racing.
[in]	max_fevals	Maximum allowed number of fevals as the additional termination condition to gen number

Exceptions

value_error if m_omega is not in the [0,1] interval, eta1, eta2 are not in the [0,1] interval, vcoeff is not in ]0,1], variant is not one of 1 .. 6, neighb_type is not one of 1 .. 4

Definition at line 58 of file pso_generational_racing.cpp.

Member Function Documentation

void pagmo::algorithm::pso_generational_racing::evolve ( population & pop ) const

virtual

Evolve implementation.

Run the PSO algorithm with incorporation of racing mechanism.

The main differences between this variant of PSO compared to the original PSO algorithm are as follows:

(1) Racing is invoked during the selection of best neighbour in locally connected toplogy. This is in contrast to the original PSO-generational algorithm, where the averaged fitness value w.r.t to current seed is used. Each of such race typically involves a small number of individuals, depending on the specific connection toplogy.

(2) During the update of memory best positions of each particle, racing is again used. Each of such race involves two decision vector, namely the current memory best position and the new candidate position.

(3) With the gbest scheme, a race over all individuals is used to identify the single best.

Parameters

[in,out] pop input/output pagmo::population to be evolved.

Implements pagmo::algorithm::base.

Definition at line 215 of file pso_generational_racing.cpp.

std::string pagmo::algorithm::pso_generational_racing::human_readable_extra ( ) const

protectedvirtual

Extra human readable algorithm info.

Will return a formatted string displaying the parameters of the algorithm.

Reimplemented from pagmo::algorithm::base.

Definition at line 874 of file pso_generational_racing.cpp.

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