High-dimensional extreme eigenvalue problems: low-rank tensor parametrization and optimization

DOI10.48550/arXiv.2609.30138MaRDI QIDQ7353523arXivFDO

Authors Pengfei Hao, Renfeng Peng, Yang Kuang, Bin Gao

Publication date 24 September 2026

Published in arXiv

Tldr We reformulate high-dimensional extreme eigenvalue problems through low-rank tensor formats and develop Riemannian optimization methods for bounded-rank tensors, achieving accuracy comparable to full-space eigensolvers while reducing computation time and storage.

Full work available at URL https://doi.org/10.48550/arXiv.2609.30138

arXiv classification

math.NAmath.OCphysics.comp-ph

Cites work

• A priori error analysis of linear and nonlinear periodic Schrödinger equations with analytic potentials
• An adjoint-based super-convergent Galerkin approximation of eigenvalues
• Calculating vibrational spectra of molecules using tensor train decomposition
• Calculating vibrational spectra of molecules using tensor train decomposition
• Computation of extreme eigenvalues in higher dimensions using block tensor train format
• Desingularization of bounded-rank matrix sets
• Desingularization of bounded-rank tensor sets
• DMRG approach to fast linear algebra in the TT-format
• Estimating a few extreme singular values and vectors for large-scale matrices in tensor train format
• Geometric Methods on Low-Rank Matrix and Tensor Manifolds
• Hearing the triangles: a numerical perspective
• Inexact subspace projection methods for low-rank tensor eigenvalue problems
• Large-scale complex eigenvalue problems
• Large-scale eigenvalue calculations for stability analysis of steady flows on massively parallel computers
• Low-rank eigenvalue solvers for block-sparse matrix product states
• Low-rank optimization on Tucker tensor varieties
• Low-rank Riemannian eigensolver for high-dimensional Hamiltonians
• Low-rank tensor methods with subspace correction for symmetric eigenvalue problems
• Multiparty entanglement in graph states
• New Riemannian preconditioned algorithms for tensor completion via polyadic decomposition
• On manifolds of tensors of fixed TT-rank
• Optimization over bounded-rank matrices through a desingularization enables joint global and local guarantees
• Preconditioned low-rank methods for high-dimensional elliptic PDE eigenvalue problems
• PRIMME: preconditioned iterative multimethod eigensolver -- methods and software description
• Riemannian optimization for high-dimensional tensor completion
• Riemannian preconditioned algorithms for tensor completion via tensor ring decomposition
• Tangent cones to tensor train varieties
• Tensor conjugate-gradient-type method for Rayleigh quotient minimization in block QTT-format
• Tensor-train decomposition
• The alternating linear scheme for tensor optimization in the tensor train format
• The density-matrix renormalization group in the age of matrix product states
• The geometry of algorithms using hierarchical tensors
• The Gross-Pitaevskii equation and eigenvector nonlinearities: numerical methods and algorithms
• Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method

This page was built for publication: High-dimensional extreme eigenvalue problems: low-rank tensor parametrization and optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7353523)