On the convergence of orthogonalization-free conjugate gradient method for extreme eigenvalues of Hermitian matrices: a Riemannian optimization interpretation
DOI10.1016/J.CAM.2024.116053MaRDI QIDQ6582011FDOQ6582011
Authors: Shixin Zheng, Haizhao Yang, Xiangxiong Zhang 
Publication date: 1 August 2024
Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)
conjugate gradientHermitian matricesRiemannian optimizationextreme eigenvaluesquotient manifoldBures-Wasserstein metricorthogonalization free
Numerical linear algebra (65Fxx) Mathematical programming (90Cxx) Numerical methods for mathematical programming, optimization and variational techniques (65Kxx)
Cites Work
- CIRR: a Rayleigh-Ritz method with contour integral for generalized eigenvalue problems
- Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method
- Numerical Optimization
- Title not available (Why is that?)
- Title not available (Why is that?)
- Introduction to Smooth Manifolds
- Global Convergence Properties of Conjugate Gradient Methods for Optimization
- Geometric diffusions as a tool for harmonic analysis and structure definition of data: diffusion maps
- A new, globally convergent Riemannian conjugate gradient method
- Self-consistent-field calculations using Chebyshev-filtered subspace iteration
- Basis selection in LOBPCG
- An efficient Gauss-Newton algorithm for symmetric low-rank product matrix approximations
- A geometric theory for preconditioned inverse iteration IV: On the fastest convergence cases
- Preconditioning orbital minimization method for planewave discretization
- A cubic scaling algorithm for excited states calculations in particle-particle random phase approximation
- A thick-restart Lanczos algorithm with polynomial filtering for Hermitian eigenvalue problems
- Computing partial spectra with least-squares rational filters
- A fast divide-and-conquer algorithm for computing the spectra of real symmetric tridiagonal matrices
- First-order methods almost always avoid strict saddle points
- Quotient geometry with simple geodesics for the manifold of fixed-rank positive-semidefinite matrices
- Coordinatewise descent methods for leading eigenvalue problem
- A robust and efficient implementation of LOBPCG
- Triangularized orthogonalization-free method for solving extreme eigenvalue problems
- Curvature of the manifold of fixed-rank positive-semidefinite matrices endowed with the Bures-Wasserstein metric
- A fast contour-integral eigensolver for non-Hermitian matrices
- A distributed block Chebyshev-Davidson algorithm for parallel spectral clustering
- Global convergence of triangularized orthogonalization-free method
- Coordinate Descent Without Coordinates: Tangent Subspace Descent on Riemannian Manifolds
This page was built for publication: On the convergence of orthogonalization-free conjugate gradient method for extreme eigenvalues of Hermitian matrices: a Riemannian optimization interpretation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6582011)
