A Cascadic Multigrid Algorithm for Computing the Fiedler Vector of Graph Laplacians

Year:    2015

Author:    John C. Urschel, Jinchao Xu, Xiaozhe Hu, Ludmil T. Zikatanov

Journal of Computational Mathematics, Vol. 33 (2015), Iss. 2 : pp. 209–226

Abstract

In this paper, we develop a cascadic multigrid algorithm for fast computation of the Fiedler vector of a graph Laplacian, namely, the eigenvector corresponding to the second smallest eigenvalue. This vector has been found to have applications in fields such as graph partitioning and graph drawing. The algorithm is a purely algebraic approach based on a heavy edge coarsening scheme and pointwise smoothing for refinement. To gain theoretical insight, we also consider the related cascadic multigrid method in the geometric setting for elliptic eigenvalue problems and show its uniform convergence under certain assumptions. Numerical tests are presented for computing the Fiedler vector of several practical graphs, and numerical results show the efficiency and optimality of our proposed cascadic multigrid algorithm.

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/jcm.1412-m2014-0041

Journal of Computational Mathematics, Vol. 33 (2015), Iss. 2 : pp. 209–226

Published online:    2015-01

AMS Subject Headings:   

Copyright:    COPYRIGHT: © Global Science Press

Pages:    18

Keywords:    Graph Laplacian Cascadic multigrid Fiedler vector Elliptic eigenvalue problems.

Author Details

John C. Urschel

Jinchao Xu

Xiaozhe Hu

Ludmil T. Zikatanov

  1. Spectral Graph Partitioning Using Geodesic Distance-based Projection

    Futamura, Yasunori | Wakaki, Ryota | Sakurai, Tetsuya

    2021 IEEE High Performance Extreme Computing Conference (HPEC), (2021), P.1

    https://doi.org/10.1109/HPEC49654.2021.9622831 [Citations: 0]
  2. On the maximal error of spectral approximation of graph bisection

    Urschel, John C. | Zikatanov, Ludmil T.

    Linear and Multilinear Algebra, Vol. 64 (2016), Iss. 10 P.1972

    https://doi.org/10.1080/03081087.2015.1133557 [Citations: 5]
  3. Graphs with absorption: Numerical methods for the absorption inverse and the computation of centrality measures

    Benzi, Michele | Fika, Paraskevi | Mitrouli, Marilena

    Linear Algebra and its Applications, Vol. 574 (2019), Iss. P.123

    https://doi.org/10.1016/j.laa.2019.03.026 [Citations: 4]
  4. On the convergence of an extrapolation cascadic multigrid method for elliptic problems

    Hu, Hongling | Ren, Zhengyong | He, Dongdong | Pan, Kejia

    Computers & Mathematics with Applications, Vol. 74 (2017), Iss. 4 P.759

    https://doi.org/10.1016/j.camwa.2017.05.023 [Citations: 20]
  5. Algebraic Multigrid for Least Squares Problems on Graphs with Applications to HodgeRank

    Colley, Charles | Lin, Junyuan | Hu, Xiaozhe | Aeron, Shuchin

    2017 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), (2017), P.627

    https://doi.org/10.1109/IPDPSW.2017.163 [Citations: 0]
  6. Fiedler Vector Approximation via Interacting Random Walks

    Doshi, Vishwaraj | Eun, Do Young

    Proceedings of the ACM on Measurement and Analysis of Computing Systems, Vol. 4 (2020), Iss. 1 P.1

    https://doi.org/10.1145/3379502 [Citations: 2]
  7. A Posteriori Error Estimates for Multilevel Methods for Graph Laplacians

    Hu, Xiaozhe | Wu, Kaiyi | Zikatanov, Ludmil T.

    SIAM Journal on Scientific Computing, Vol. 43 (2021), Iss. 5 P.S727

    https://doi.org/10.1137/20M1349618 [Citations: 3]
  8. An Adaptive Multigrid Method Based on Path Cover

    Hu, Xiaozhe | Lin, Junyuan | Zikatanov, Ludmil T.

    SIAM Journal on Scientific Computing, Vol. 41 (2019), Iss. 5 P.S220

    https://doi.org/10.1137/18M1194493 [Citations: 10]
  9. The Fiedler Vector of a Laplacian Tensor for Hypergraph Partitioning

    Chen, Yannan | Qi, Liqun | Zhang, Xiaoyan

    SIAM Journal on Scientific Computing, Vol. 39 (2017), Iss. 6 P.A2508

    https://doi.org/10.1137/16M1094828 [Citations: 25]
  10. Performance-Portable Graph Coarsening for Efficient Multilevel Graph Analysis

    Gilbert, Michael S. | Acer, Seher | Boman, Erik G. | Madduri, Kamesh | Rajamanickam, Sivasankaran

    2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS), (2021), P.213

    https://doi.org/10.1109/IPDPS49936.2021.00030 [Citations: 3]
  11. Maximizing Algebraic Connectivity of Constrained Graphs in Adversarial Environments

    Anderson, Tor | Chang, Chin-Yao | Martinez, Sonia

    2018 European Control Conference (ECC), (2018), P.125

    https://doi.org/10.23919/ECC.2018.8550531 [Citations: 3]
  12. Bringing physics into the coarse‐grid selection: Approximate diffusion distance/effective resistance measures for network analysis and algebraic multigrid for graph Laplacians and systems of elliptic partial differential equations

    Lee, Barry

    Numerical Linear Algebra with Applications, Vol. 31 (2024), Iss. 2

    https://doi.org/10.1002/nla.2539 [Citations: 1]
  13. Finding diverse ways to improve algebraic connectivity through multi-start optimization

    Mackay, Sarah | Ponce, Colin | Osborn, Sarah | McGarry, Meghan | Pinar, Ali

    Journal of Complex Networks, Vol. 9 (2021), Iss. 1

    https://doi.org/10.1093/comnet/cnab005 [Citations: 1]
  14. A Bootstrap Multigrid Eigensolver

    Brannick, James | Cao, Shuhao

    SIAM Journal on Matrix Analysis and Applications, Vol. 43 (2022), Iss. 4 P.1627

    https://doi.org/10.1137/20M131151X [Citations: 1]
  15. Computing the diffusion state distance on graphs via algebraic multigrid and random projections

    Lin, Junyuan | Cowen, Lenore J. | Hescott, Benjamin | Hu, Xiaozhe

    Numerical Linear Algebra with Applications, Vol. 25 (2018), Iss. 3

    https://doi.org/10.1002/nla.2156 [Citations: 7]
  16. On the approximation of Laplacian eigenvalues in graph disaggregation

    Hu, Xiaozhe | Urschel, John C. | Zikatanov, Ludmil T.

    Linear and Multilinear Algebra, Vol. 65 (2017), Iss. 9 P.1805

    https://doi.org/10.1080/03081087.2016.1256944 [Citations: 1]