MatCreateLMVMDiagBroyden#

DiagBrdn creates a symmetric Broyden-type diagonal matrix used for approximating Hessians.

Synopsis#

#include "petscksp.h" 
PetscErrorCode MatCreateLMVMDiagBroyden(MPI_Comm comm, PetscInt n, PetscInt N, Mat *B)

Collective

Input Parameters#

  • comm - MPI communicator

  • n - number of local rows for storage vectors

  • N - global size of the storage vectors

Output Parameter#

  • B - the matrix

Options Database Keys#

  • -mat_lmvm_theta - (developer) convex ratio between BFGS and DFP components of the diagonal J0 scaling

  • -mat_lmvm_rho - (developer) update limiter for the J0 scaling

  • -mat_lmvm_alpha - (developer) coefficient factor for the quadratic subproblem in J0 scaling

  • -mat_lmvm_beta - (developer) exponential factor for the diagonal J0 scaling

  • -mat_lmvm_sigma_hist - (developer) number of past updates to use in J0 scaling.

  • -mat_lmvm_tol - (developer) tolerance for bounding the denominator of the rescaling away from 0.

  • -mat_lmvm_forward - (developer) whether or not to use the forward or backward Broyden update to the diagonal

Notes#

It is recommended that one use the MatCreate(), MatSetType() and/or MatSetFromOptions() paradigm instead of this routine directly.

It consists of a convex combination of DFP and BFGS diagonal approximation schemes, such that DiagBrdn = (1-theta)BFGS + thetaDFP. To preserve symmetric positive-definiteness, we restrict theta to be in [0, 1]. We also ensure positive definiteness by taking the VecAbs() of the final vector.

There are two ways of approximating the diagonal: using the forward (B) update schemes for BFGS and DFP and then taking the inverse, or directly working with the inverse (H) update schemes for the BFGS and DFP updates, derived using the Sherman-Morrison-Woodbury formula. We have implemented both, controlled by a parameter below.

In order to use the DiagBrdn matrix with other vector types, i.e. doing matrix-vector products and matrix solves, the matrix must first be created using MatCreate() and MatSetType(), followed by MatLMVMAllocate(). Then it will be available for updating (via MatLMVMUpdate()) in one’s favored solver implementation.

See Also#

KSP: Linear System Solvers, MatCreate(), MATLMVM, MATLMVMDIAGBRDN, MatCreateLMVMDFP(), MatCreateLMVMSR1(), MatCreateLMVMBFGS(), MatCreateLMVMBrdn(), MatCreateLMVMSymBrdn()

Level#

intermediate

Location#

src/ksp/ksp/utils/lmvm/diagbrdn/diagbrdn.c

Index of all KSP routines
Table of Contents for all manual pages
Index of all manual pages