A modification of the Jacobi-Davidson method

Mashetti Ravibabu

Home/Archive/arXiv

arXiv · 1902.02285

A modification of the Jacobi-Davidson method

Mashetti Ravibabu

View original source ↗Read source document ↗Export metadata ↓

Abstract

Each iteration in Jacobi-Davidson method for solving large sparse eigenvalue problems involves two phases, called subspace expansion and eigen pair extraction. The subspace expansion phase involves solving a correction equation. We propose a modification to this by introducing a related correction equation, motivated by the least squares. We call the proposed method as the Modified Jacobi-Davidson Method. When the subspace expansion is ignored as in the Simplified Jacobi- Davidson Method, the modified method is called as Modified Simplified Jacobi-Davidson Method. We analyze the convergence properties of the proposed method for Symmetric matrices. Numerical experiments have been carried out to check whether the method is computationally viable or not.

Explore related subjects

math.NA cs.NA

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Mashetti Ravibabu. 2019-02-06. A modification of the Jacobi-Davidson method. https://arxiv.org/abs/1902.02285

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Structure-Aware Analyses and Algorithms for Interpolative Decompositions

Low-rank approximation is a task of critical importance in modern science, engineering, and statistics. Many low-rank approximation algorithms, such as the randomized singular value decomposition (RSVD), project their input matrix into a subspace approximating the span of its leading singular vectors. Other algorithms compress their input into a small subset of representative rows or columns, leading to a so-called interpolative decomposition. This paper investigates how the accuracy of interpolative decompositions is affected by the structural properties of the input matrix being operated on, including how these properties affect the performance comparison between interpolative decompositions and RSVD. We give particular focus to a randomized Golub-Klema-Stewart (RGKS) algorithm, an interpolative decomposition algorithm which combines RSVD with the Golub-Klema-Stewart (GKS) pivoting strategy. Through numerical experiments, we find that matrix structures including singular subspace geometry and singular spectrum decay play a significant role in determining the performance comparison between these different algorithms. We also prove inequalities which bound the error of a general interpolative decomposition in terms of these matrix structures. Finally, we develop forms of these bounds specialized to RGKS while considering how randomization affects the approximation error of this algorithm.

math.NA↗

arXiv2026

Integral equations for flexural-gravity waves: analysis and numerical methods

In this work, we develop a fast and accurate method for the scattering of flexural-gravity waves by a thin plate of varying thickness overlying a fluid of infinite depth. This problem commonly arises in the study of sea ice and ice shelves, which can have complicated heterogeneities that include ridges and rolls. With certain natural assumptions on the thickness, we present an integral equation formulation for solving this class of problems and analyze its mathematical properties. The integral equation is then discretized and solved using a high-order-accurate, FFT-accelerated algorithm. The speed, accuracy, and scalability of this approach are demonstrated through a variety of illustrative examples.

math.NA↗

arXiv2026

Maximum bound principle for Q-tensor gradient flow with low regularity integrators

The Landau-de Gennes (LdG) theory is a widely used thermodynamic continuum framework for modeling the behavior of ordered states and defects in liquid crystals with a tensor-order parameter $Q$. In this study, we develop and analyze first- and second-order low-regularity integrator (LRI) schemes for the $Q$-tensor gradient flow and prove the maximum bound principle. In particular, through the reformulation of the LRI schemes, we establish rigorous modified energy dissipation laws for the LRI1a and LRI1b schemes, thereby filling a significant theoretical gap in the existing literature on LRI methods. Moreover, this reformulation establishes a structural bridge between the LRI schemes and backward differentiation formula (BDF) methods, which opens up new possibilities for the construction and analysis of LRI-type methods. We then establish first- and second-order temporal convergence under $H^1$ and $H^2$ regularity assumptions, respectively. Several numerical experiments are presented to validate our theoretical results and to simulate the evolution of defect dynamics.

math.NA↗