ROL
Public Member Functions | Private Attributes | List of all members
ROL::PQNObjective< Real > Class Template Reference

Provides the interface to evaluate the quadratic quasi-Newton objective. More...

#include <ROL_PQNObjective.hpp>

+ Inheritance diagram for ROL::PQNObjective< Real >:

Public Member Functions

 PQNObjective (const Ptr< Secant< Real >> &secant, const Vector< Real > &x, const Vector< Real > &g)
 
Real value (const Vector< Real > &x, Real &tol) override
 Compute value. More...
 
void gradient (Vector< Real > &g, const Vector< Real > &x, Real &tol) override
 Compute gradient. More...
 
void hessVec (Vector< Real > &hv, const Vector< Real > &v, const Vector< Real > &x, Real &tol) override
 Apply Hessian approximation to vector. More...
 
void setAnchor (const Vector< Real > &x, const Vector< Real > &g)
 
- Public Member Functions inherited from ROL::Objective< Real >
virtual ~Objective ()
 
 Objective ()
 
virtual void update (const Vector< Real > &x, UpdateType type, int iter=-1)
 Update objective function. More...
 
virtual void update (const Vector< Real > &x, bool flag=true, int iter=-1)
 Update objective function. More...
 
virtual Real dirDeriv (const Vector< Real > &x, const Vector< Real > &d, Real &tol)
 Compute directional derivative. More...
 
virtual void invHessVec (Vector< Real > &hv, const Vector< Real > &v, const Vector< Real > &x, Real &tol)
 Apply inverse Hessian approximation to vector. More...
 
virtual void precond (Vector< Real > &Pv, const Vector< Real > &v, const Vector< Real > &x, Real &tol)
 Apply preconditioner to vector. More...
 
virtual void prox (Vector< Real > &Pv, const Vector< Real > &v, Real t, Real &tol)
 
virtual std::vector
< std::vector< Real > > 
checkGradient (const Vector< Real > &x, const Vector< Real > &d, const bool printToStream=true, std::ostream &outStream=std::cout, const int numSteps=ROL_NUM_CHECKDERIV_STEPS, const int order=1)
 Finite-difference gradient check. More...
 
virtual std::vector
< std::vector< Real > > 
checkGradient (const Vector< Real > &x, const Vector< Real > &g, const Vector< Real > &d, const bool printToStream=true, std::ostream &outStream=std::cout, const int numSteps=ROL_NUM_CHECKDERIV_STEPS, const int order=1)
 Finite-difference gradient check. More...
 
virtual std::vector
< std::vector< Real > > 
checkGradient (const Vector< Real > &x, const Vector< Real > &d, const std::vector< Real > &steps, const bool printToStream=true, std::ostream &outStream=std::cout, const int order=1)
 Finite-difference gradient check with specified step sizes. More...
 
virtual std::vector
< std::vector< Real > > 
checkGradient (const Vector< Real > &x, const Vector< Real > &g, const Vector< Real > &d, const std::vector< Real > &steps, const bool printToStream=true, std::ostream &outStream=std::cout, const int order=1)
 Finite-difference gradient check with specified step sizes. More...
 
virtual std::vector
< std::vector< Real > > 
checkHessVec (const Vector< Real > &x, const Vector< Real > &v, const bool printToStream=true, std::ostream &outStream=std::cout, const int numSteps=ROL_NUM_CHECKDERIV_STEPS, const int order=1)
 Finite-difference Hessian-applied-to-vector check. More...
 
virtual std::vector
< std::vector< Real > > 
checkHessVec (const Vector< Real > &x, const Vector< Real > &hv, const Vector< Real > &v, const bool printToStream=true, std::ostream &outStream=std::cout, const int numSteps=ROL_NUM_CHECKDERIV_STEPS, const int order=1)
 Finite-difference Hessian-applied-to-vector check. More...
 
virtual std::vector
< std::vector< Real > > 
checkHessVec (const Vector< Real > &x, const Vector< Real > &v, const std::vector< Real > &steps, const bool printToStream=true, std::ostream &outStream=std::cout, const int order=1)
 Finite-difference Hessian-applied-to-vector check with specified step sizes. More...
 
virtual std::vector
< std::vector< Real > > 
checkHessVec (const Vector< Real > &x, const Vector< Real > &hv, const Vector< Real > &v, const std::vector< Real > &steps, const bool printToStream=true, std::ostream &outStream=std::cout, const int order=1)
 Finite-difference Hessian-applied-to-vector check with specified step sizes. More...
 
virtual std::vector< Real > checkHessSym (const Vector< Real > &x, const Vector< Real > &v, const Vector< Real > &w, const bool printToStream=true, std::ostream &outStream=std::cout)
 Hessian symmetry check. More...
 
virtual std::vector< Real > checkHessSym (const Vector< Real > &x, const Vector< Real > &hv, const Vector< Real > &v, const Vector< Real > &w, const bool printToStream=true, std::ostream &outStream=std::cout)
 Hessian symmetry check. More...
 
virtual void setParameter (const std::vector< Real > &param)
 

Private Attributes

const Ptr< Secant< Real > > secant_
 
const Ptr< Vector< Real > > x_
 
const Ptr< Vector< Real > > g_
 
const Ptr< Vector< Real > > pwa_
 
const Ptr< Vector< Real > > dwa_
 

Additional Inherited Members

- Protected Member Functions inherited from ROL::Objective< Real >
const std::vector< Real > getParameter (void) const
 

Detailed Description

template<typename Real>
class ROL::PQNObjective< Real >

Provides the interface to evaluate the quadratic quasi-Newton objective.

This class implements the PQN quasi-Newton objective for use with ROL::TypeB::QuasiNewtonAlgorithm. Given a function \(f:\mathcal{X}\to\mathbb{R}\) and a Hessian approximation \(B_k:\mathcal{X}}\to\mathcal{X}^*\), the functional is

\[ q_k(x) = \frac{1}{2}\langle B_k (x-x_k),(x-x_k)\rangle_{\mathcal{X}^*,\mathcal{X}} + \langle f'(x_k), (x-x_k)\rangle_{\mathcal{X}^*,\mathcal{X}}. \]


Definition at line 70 of file ROL_PQNObjective.hpp.

Constructor & Destructor Documentation

template<typename Real >
ROL::PQNObjective< Real >::PQNObjective ( const Ptr< Secant< Real >> &  secant,
const Vector< Real > &  x,
const Vector< Real > &  g 
)

Definition at line 51 of file ROL_PQNObjective_Def.hpp.

References ROL::PQNObjective< Real >::setAnchor().

Member Function Documentation

template<typename Real >
Real ROL::PQNObjective< Real >::value ( const Vector< Real > &  x,
Real &  tol 
)
overridevirtual

Compute value.

This function returns the objective function value.

Parameters
[in]xis the current iterate.
[in]tolis a tolerance for inexact objective function computation.

Implements ROL::Objective< Real >.

Definition at line 59 of file ROL_PQNObjective_Def.hpp.

template<typename Real >
void ROL::PQNObjective< Real >::gradient ( Vector< Real > &  g,
const Vector< Real > &  x,
Real &  tol 
)
overridevirtual

Compute gradient.

This function returns the objective function gradient.

Parameters
[out]gis the gradient.
[in]xis the current iterate.
[in]tolis a tolerance for inexact objective function computation.

The default implementation is a finite-difference approximation based on the function value. This requires the definition of a basis \(\{\phi_i\}\) for the optimization vectors x and the definition of a basis \(\{\psi_j\}\) for the dual optimization vectors (gradient vectors g). The bases must be related through the Riesz map, i.e., \( R \{\phi_i\} = \{\psi_j\}\), and this must be reflected in the implementation of the ROL::Vector::dual() method.

Reimplemented from ROL::Objective< Real >.

Definition at line 69 of file ROL_PQNObjective_Def.hpp.

References ROL::Vector< Real >::plus().

template<typename Real >
void ROL::PQNObjective< Real >::hessVec ( Vector< Real > &  hv,
const Vector< Real > &  v,
const Vector< Real > &  x,
Real &  tol 
)
overridevirtual

Apply Hessian approximation to vector.

This function applies the Hessian of the objective function to the vector \(v\).

Parameters
[out]hvis the the action of the Hessian on \(v\).
[in]vis the direction vector.
[in]xis the current iterate.
[in]tolis a tolerance for inexact objective function computation.

Reimplemented from ROL::Objective< Real >.

Definition at line 77 of file ROL_PQNObjective_Def.hpp.

template<typename Real >
void ROL::PQNObjective< Real >::setAnchor ( const Vector< Real > &  x,
const Vector< Real > &  g 
)

Definition at line 82 of file ROL_PQNObjective_Def.hpp.

Referenced by ROL::PQNObjective< Real >::PQNObjective().

Member Data Documentation

template<typename Real >
const Ptr<Secant<Real> > ROL::PQNObjective< Real >::secant_
private

Definition at line 72 of file ROL_PQNObjective.hpp.

template<typename Real >
const Ptr<Vector<Real> > ROL::PQNObjective< Real >::x_
private

Definition at line 73 of file ROL_PQNObjective.hpp.

template<typename Real >
const Ptr<Vector<Real> > ROL::PQNObjective< Real >::g_
private

Definition at line 73 of file ROL_PQNObjective.hpp.

template<typename Real >
const Ptr<Vector<Real> > ROL::PQNObjective< Real >::pwa_
private

Definition at line 73 of file ROL_PQNObjective.hpp.

template<typename Real >
const Ptr<Vector<Real> > ROL::PQNObjective< Real >::dwa_
private

Definition at line 73 of file ROL_PQNObjective.hpp.


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