BOLT: Block-Orthonormal Lanczos for Trace estimation of matrix functions

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yeon, Kingsley, Ghosal, Promit, Anitescu, Mihai
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908720604119040
author Yeon, Kingsley
Ghosal, Promit
Anitescu, Mihai
author_facet Yeon, Kingsley
Ghosal, Promit
Anitescu, Mihai
contents Efficient matrix trace estimation is essential for scalable computation of log-determinants, matrix norms, and distributional divergences. In many large-scale applications, the matrices involved are too large to store or access in full, making even a single matrix-vector (mat-vec) product infeasible. Instead, one often has access only to small subblocks of the matrix or localized matrix-vector products on restricted index sets. Hutch++ achieves optimal convergence rate but relies on randomized SVD and assumes full mat-vec access, making it difficult to apply in these constrained settings. We propose the Block-Orthonormal Stochastic Lanczos Quadrature (BOLT), which matches Hutch++ accuracy with a simpler implementation based on orthonormal block probes and Lanczos iterations. BOLT builds on the Stochastic Lanczos Quadrature (SLQ) framework, which combines random probing with Krylov subspace methods to efficiently approximate traces of matrix functions, and performs better than Hutch++ in near flat-spectrum regimes. To address memory limitations and partial access constraints, we introduce Subblock SLQ, a variant of BOLT that operates only on small principal submatrices. As a result, this framework yields a proxy KL divergence estimator and an efficient method for computing the Wasserstein-2 distance between Gaussians - both compatible with low-memory and partial-access regimes. We provide theoretical guarantees and demonstrate strong empirical performance across a range of high-dimensional settings.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12289
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BOLT: Block-Orthonormal Lanczos for Trace estimation of matrix functions
Yeon, Kingsley
Ghosal, Promit
Anitescu, Mihai
Numerical Analysis
Data Structures and Algorithms
Machine Learning
65F60, 65F10, 65C60, 68W20
Efficient matrix trace estimation is essential for scalable computation of log-determinants, matrix norms, and distributional divergences. In many large-scale applications, the matrices involved are too large to store or access in full, making even a single matrix-vector (mat-vec) product infeasible. Instead, one often has access only to small subblocks of the matrix or localized matrix-vector products on restricted index sets. Hutch++ achieves optimal convergence rate but relies on randomized SVD and assumes full mat-vec access, making it difficult to apply in these constrained settings. We propose the Block-Orthonormal Stochastic Lanczos Quadrature (BOLT), which matches Hutch++ accuracy with a simpler implementation based on orthonormal block probes and Lanczos iterations. BOLT builds on the Stochastic Lanczos Quadrature (SLQ) framework, which combines random probing with Krylov subspace methods to efficiently approximate traces of matrix functions, and performs better than Hutch++ in near flat-spectrum regimes. To address memory limitations and partial access constraints, we introduce Subblock SLQ, a variant of BOLT that operates only on small principal submatrices. As a result, this framework yields a proxy KL divergence estimator and an efficient method for computing the Wasserstein-2 distance between Gaussians - both compatible with low-memory and partial-access regimes. We provide theoretical guarantees and demonstrate strong empirical performance across a range of high-dimensional settings.
title BOLT: Block-Orthonormal Lanczos for Trace estimation of matrix functions
topic Numerical Analysis
Data Structures and Algorithms
Machine Learning
65F60, 65F10, 65C60, 68W20
url https://arxiv.org/abs/2505.12289