A POSTERIORI ERROR ESTIMATES FOR MULTILEVEL METHODS FOR GRAPH LAPLACIANS

被引:4
作者
Hu, Xiaozhe [1 ]
Wu, Kaiyi [1 ]
Zikatanov, Ludmil T. [2 ]
机构
[1] Tufts Univ, Dept Math, Medford, MA 02155 USA
[2] Penn State Univ, Dept Math, University Pk, PA 16802 USA
基金
美国国家科学基金会;
关键词
graph Laplacian; a posteriori error estimates; cycle space; spanning tree; Helmholtz decomposition; AGGREGATION ALPHA-SA; MULTIGRID METHOD; CONVERGENCE; NETWORKS; VECTOR; BOUNDS;
D O I
10.1137/20M1349618
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
In this paper, we study a posteriori error estimators which aid multilevel iterative solvers for linear systems of graph Laplacians. In earlier works such estimates were computed by solving a perturbed global optimization problem, which could be computationally expensive. We propose a novel strategy to compute these estimates by constructing a Helmholtz decomposition on the graph based on a spanning tree and the corresponding cycle space. To compute the error estimator, we solve a linear system efficiently on the spanning tree and then a least-squares problem on the cycle space. As we show, such an estimator has a nearly linear computational complexity for sparse graphs under certain assumptions. Numerical experiments are presented to demonstrate the efficacy of the proposed method.
引用
收藏
页码:S727 / S742
页数:16
相关论文
共 64 条
[21]  
Cormen T. H., 2009, INTRO ALGORITHMS, V3rd
[22]   Adaptive AMG with coarsening based on compatible weighted matching [J].
D'Ambra, Pasqua ;
Vassilevski, Panayot S. .
COMPUTING AND VISUALIZATION IN SCIENCE, 2013, 16 (02) :59-76
[23]   Explicit error bounds in a conforming finite element method [J].
Destuynder, P ;
Métivet, B .
MATHEMATICS OF COMPUTATION, 1999, 68 (228) :1379-1396
[24]   Random-walk computation of similarities between nodes of a graph with application to collaborative recommendation [J].
Fouss, Francois ;
Pirotte, Alain ;
Renders, Jean-Michel ;
Saerens, Marco .
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2007, 19 (03) :355-369
[25]   Semisupervised image classification with Laplacian support vector machines [J].
Gomez-Chova, Luis ;
Camps-Valls, Gustavo ;
Munoz-Mari, Jordi ;
Calpe, Javier .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2008, 5 (03) :336-340
[26]   GRAPH THEORY AND MOLECULAR-ORBITALS - TOTAL PI-ELECTRON ENERGY OF ALTERNANT HYDROCARBONS [J].
GUTMAN, I ;
TRINAJSTIC, N .
CHEMICAL PHYSICS LETTERS, 1972, 17 (04) :535-538
[27]  
Hamilton W. L., 2017, Representation Learning on Graphs: methods and Applications
[28]   AN ADAPTIVE MULTIGRID METHOD BASED ON PATH COVER [J].
Hu, Xiaozhe ;
Lin, Junyuan ;
Zikatanov, Ludmil T. .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2019, 41 (05) :S220-S241
[29]   Cycle bases in graphs characterization, algorithms, complexity, and applications [J].
Kavitha, Telikepalli ;
Liebchen, Christian ;
Mehlhorn, Kurt ;
Michail, Dimitrios ;
Rizzi, Romeo ;
Ueckerdt, Torsten ;
Zwei, Katharina A. .
COMPUTER SCIENCE REVIEW, 2009, 3 (04) :199-243
[30]   A-POSTERIORI ERROR ANALYSIS AND ADAPTIVE PROCESSES IN THE FINITE-ELEMENT METHOD .1. ERROR ANALYSIS [J].
KELLY, DW ;
GAGO, JPDR ;
ZIENKIEWICZ, OC ;
BABUSKA, I .
INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 1983, 19 (11) :1593-1619