AndersonAccel< Conf > Class Template Reference

#include <alpaqa/accelerators/anderson.hpp>

Detailed Description

template<Config Conf = DefaultConfig>
class alpaqa::AndersonAccel< Conf >

Anderson Acceleration.

Algorithm for accelerating fixed-point iterations for finding fixed points of a function \( g \), i.e. \( g(x^\star) = x^\star \), or equivalently, roots of the residual \( r(x) \triangleq g(x) - x \).

Definition at line 38 of file anderson.hpp.

Collaboration diagram for AndersonAccel< Conf >:

Public Types
using	Params = AndersonAccelParams< config_t >

Public Member Functions
	AndersonAccel ()=default
	AndersonAccel (Params params)
	AndersonAccel (Params params, length_t n)
void	resize (length_t n)
	Change the problem dimension.
void	initialize (crvec g_0, crvec r_0)
	Call this function on the first iteration to initialize the accelerator.
void	compute (crvec gₖ, crvec rₖ, rvec xₖ_aa)
	Compute the accelerated iterate \( x^k_\text{AA} \), given the function value at the current iterate \( g^k = g(x^k) \) and the corresponding residual \( r^k = g^k - x^k \).
void	compute (crvec gₖ, vec &&rₖ, rvec xₖ_aa)
	Compute the accelerated iterate \( x^k_\text{AA} \), given the function value at the current iterate \( g^k = g(x^k) \) and the corresponding residual \( r^k = g^k - x^k \).
void	reset ()
	Reset the accelerator (but keep the last function value and residual, so calling initialize is not necessary).
length_t	n () const
	Get the problem dimension.
length_t	history () const
	Get the maximum number of stored columns.
length_t	current_history () const
	Get the number of columns currently stored in the buffer.
const Params &	get_params () const
	Get the parameters.
std::string	get_name () const
void	scale_R (real_t scal)
	Scale the factorization.

Private Attributes
Params	params
LimitedMemoryQR< config_t >	qr
mat	G
vec	rₗₐₛₜ
vec	γ_LS
bool	initialized = false

Member Typedef Documentation

Params

using Params = AndersonAccelParams<config_t>

Definition at line 41 of file anderson.hpp.

Constructor & Destructor Documentation

AndersonAccel() [1/3]

AndersonAccel ( )

default

AndersonAccel() [2/3]

AndersonAccel ( Params  params )

inline

Parameters

params

Parameters.

Definition at line 46 of file anderson.hpp.

AndersonAccel() [3/3]

AndersonAccel ( Params  params, length_t  n  )

inline

Parameters

params	Parameters.
n	Problem dimension (size of the vectors).

Definition at line 51 of file anderson.hpp.

Here is the call graph for this function:

Member Function Documentation

resize()

void resize ( length_t  n )

inline

Change the problem dimension.

Flushes the history.

Parameters

n	Problem dimension (size of the vectors).

Definition at line 56 of file anderson.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

initialize()

void initialize ( crvec  g_0, crvec  r_0  )

inline

Call this function on the first iteration to initialize the accelerator.

Definition at line 66 of file anderson.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

compute() [1/2]

void compute ( crvec  gₖ, crvec  rₖ, rvec  xₖ_aa  )

inline

Compute the accelerated iterate \( x^k_\text{AA} \), given the function value at the current iterate \( g^k = g(x^k) \) and the corresponding residual \( r^k = g^k - x^k \).

Definition at line 78 of file anderson.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

compute() [2/2]

void compute ( crvec  gₖ, vec &&  rₖ, rvec  xₖ_aa  )

inline

Compute the accelerated iterate \( x^k_\text{AA} \), given the function value at the current iterate \( g^k = g(x^k) \) and the corresponding residual \( r^k = g^k - x^k \).

Definition at line 88 of file anderson.hpp.

Here is the call graph for this function:

reset()

void reset ( )

inline

Reset the accelerator (but keep the last function value and residual, so calling initialize is not necessary).

Definition at line 100 of file anderson.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

n()

length_t n ( ) const

inline

Get the problem dimension.

Definition at line 108 of file anderson.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

history()

length_t history ( ) const

inline

Get the maximum number of stored columns.

Definition at line 110 of file anderson.hpp.

Here is the call graph for this function:

current_history()

length_t current_history ( ) const

inline

Get the number of columns currently stored in the buffer.

Definition at line 112 of file anderson.hpp.

Here is the call graph for this function:

get_params()

const Params & get_params ( ) const

inline

Get the parameters.

Definition at line 115 of file anderson.hpp.

Here is the caller graph for this function:

get_name()

std::string get_name ( ) const

inline

Definition at line 117 of file anderson.hpp.

scale_R()

void scale_R ( real_t  scal )

inline

Scale the factorization.

Definition at line 122 of file anderson.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

Member Data Documentation

params

Params params

private

Definition at line 125 of file anderson.hpp.

qr

LimitedMemoryQR<config_t> qr

private

Definition at line 126 of file anderson.hpp.

G

mat G

private

Definition at line 127 of file anderson.hpp.

rₗₐₛₜ

vec rₗₐₛₜ

private

Definition at line 128 of file anderson.hpp.

γ_LS

vec γ_LS

private

Definition at line 129 of file anderson.hpp.

initialized

bool initialized = false

private

Definition at line 130 of file anderson.hpp.

---

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

• src/alpaqa/include/alpaqa/accelerators/anderson.hpp