Hostname: page-component-76fb5796d-25wd4 Total loading time: 0 Render date: 2024-04-29T15:53:03.964Z Has data issue: false hasContentIssue false
  • English
  • Français

Phase transition for the generalized two-community stochastic block model

Part of: Special matrices Probability theory on algebraic and topological structures Theory of computing

Published online by Cambridge University Press:  31 July 2023

Sunmin Lee Open the ORCID record for Sunmin Lee [Opens in a new window]  and
Ji Oon Lee
Sunmin Lee*
Affiliation:
Korea Advanced Institute of Science and Technology (KAIST)
Ji Oon Lee*
Affiliation:
Korea Advanced Institute of Science and Technology (KAIST)
*
*Postal address: Department of Mathematical Sciences, KAIST, Daejeon 34141, South Korea.
*Postal address: Department of Mathematical Sciences, KAIST, Daejeon 34141, South Korea.
Get access Rights & Permissions [Opens in a new window]

Abstract

We study the problem of detecting the community structure from the generalized stochastic block model with two communities (G2-SBM). Based on analysis of the Stieljtes transform of the empirical spectral distribution, we prove a Baik–Ben Arous–Péché (BBP)-type transition for the largest eigenvalue of the G2-SBM. For specific models, such as a hidden community model and an unbalanced stochastic block model, we provide precise formulas for the two largest eigenvalues, establishing the gap in the BBP-type transition.

Keywords

BBP-type transitionempirical spectral distributionWigner-type matrixhidden community modelunbalanced stochastic block model

MSC classification

Primary: 68Q87: Probability in computer science (algorithm analysis, random structures, phase transitions, etc.) 15B51: Stochastic matrices
Secondary: 60B20: Random matrices (probabilistic aspects; for algebraic aspects see )
Type
Original Article
Information
Journal of Applied Probability , First View , pp. 1 - 16
Check for updates
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of Applied Probability Trust

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Abbe, E. (2018). Community detection and stochastic block models: Recent developments. J. Mach. Learn. Res. 18, 1–86.Google Scholar
Abbe, E., Bandeira, A. S. and Hall, G. (2016). Exact recovery in the stochastic block model. IEEE Trans. Inf. Theory 62, 471–487.CrossRefGoogle Scholar
Abbe, E. and Sandon, C. (2016). Crossing the KS threshold in the stochastic block model with information theory. In 2016 IEEE International Symposium on Information Theory (ISIT). Barcelona, Spain. pp. 840–844.CrossRefGoogle Scholar
Ajanki, O., Erdős, L. and Krüger, T. (2017). Universality for general Wigner-type matrices. Prob. Theory Relat. Fields 169, 667–727.CrossRefGoogle Scholar
Ajanki, O., Erdős, L. and Krüger, T. (2019). Quadratic Vector Equations on Complex Upper Half-Plane (Memoirs of the American Mathematical Society 261). American Mathematical Society, Providence, RI.Google Scholar
Baik, J., Ben Arous, G. and Péché, S. (2005). Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices. Ann. Prob. 33, 1643–1697.CrossRefGoogle Scholar
Banks, J., Moore, C., Neeman, J. and Netrapalli, P. (2016). Information-theoretic thresholds for community detection in sparse networks. Proc. Mach. Learn. Res. 49, 383–416.Google Scholar
Barbier, J., Dia, M., Macris, N., Krzakala, F., Lesieur, T. and Zdeborová, L. (2016). Mutual information for symmetric rank-one matrix estimation: A proof of the replica formula. In Advances in Neural Information Processing Systems 29 (NIPS 2016), eds D. Lee, M. Sugiyama, U. Luxburg, I. Guyon and R. Garnett. Curran Associates, Inc., Barcelona.Google Scholar
Benaych-Georges, F. and Nadakuditi, R. R. (2011). The eigenvalues and eigenvectors of finite, low rank perturbations of large random matrices. Adv. Math. 227, 494–521.CrossRefGoogle Scholar
Capitaine, M., Donati-Martin, C. and Féral, D. (2009). The largest eigenvalues of finite rank deformation of large Wigner matrices: Convergence and nonuniversality of the fluctuations. Ann. Prob. 37, 1–47.CrossRefGoogle Scholar
Chen, P.-Y. and Hero, A. (2015). Universal phase transition in community detectability under a stochastic block model. Phys. Rev. E 91, 032804.CrossRefGoogle Scholar
Chung, H. W. and Lee, J. O. (2019). Weak detection of signal in the spiked Wigner model. Proc. Mach. Learn. Res. 97, 1233–1241.Google Scholar
Dumitriu, I. and Zhu, Y. (2019). Sparse general Wigner-type matrices: Local law and eigenvector delocalization. J. Math. Phys. 60, 023301.CrossRefGoogle Scholar
El Alaoui, A., Krzakala, F. and Jordan, M. I. (2020). Fundamental limits of detection in the spiked Wigner model. Ann. Statist. 48, 863–885.CrossRefGoogle Scholar
Erdös, L., Krüger, T. and Schröder, D. (2020). Cusp universality for random matrices I: Local law and the complex Hermitian case. Commun. Math. Phys. 378, 1203–1278.CrossRefGoogle ScholarPubMed
Erdös, L. and Mühlbacher, P. (2019). Bounds on the norm of Wigner-type random matrices. Random Matrices Theory Appl. 8, 1950009.CrossRefGoogle Scholar
Féral, D. and Péché, S. (2007). The largest eigenvalue of rank one deformation of large Wigner matrices. Commun. Math. Phys. 272, 185–228.CrossRefGoogle Scholar
Hajek, B., Wu, Y. and Xu, J. (2016). Achieving exact cluster recovery threshold via semidefinite programming. IEEE Trans. Inf. Theory 62, 2788–2797.CrossRefGoogle Scholar
Hajek, B., Wu, Y. and Xu, J. (2017). Information limits for recovering a hidden community. IEEE Trans. Inf. Theory 63, 4729–4745.CrossRefGoogle Scholar
Hajek, B., Wu, Y. and Xu, J. (2018). Recovering a hidden community beyond the Kesten–Stigum threshold in $o (| e| \log^*|v|)$ time. J. Appl. Prob. 55, 325–352.CrossRefGoogle Scholar
Jung, J. H., Chung, H. W. and Lee, J. O. (2020). Weak detection in the spiked Wigner model with general rank. Preprint, arXiv:2001.05676.Google Scholar
Mossel, E., Neeman, J. and Sly, A. (2018). A proof of the block model threshold conjecture. Combinatorica 38, 665–708.CrossRefGoogle Scholar
Mossel, E. and Xu, J. (2016). Density evolution in the degree-correlated stochastic block model. Proc. Mach. Learn. Res. 49, 1319–1356.Google Scholar
Péché, S. (2006). The largest eigenvalue of small rank perturbations of Hermitian random matrices. Prob. Theory Relat. Fields 134, 127–173.CrossRefGoogle Scholar
Perry, A., Wein, A. S., Bandeira, A. S. and Moitra, A. (2018). Optimality and sub-optimality of PCA I: Spiked random matrix models. Ann. Statist. 46, 2416–2451.CrossRefGoogle Scholar
Stanley, N., Bonacci, T., Kwitt, R., Niethammer, M. and Mucha, P. J. (2019). Stochastic block models with multiple continuous attributes. Appl. Network Sci. 4, 1–22.CrossRefGoogle Scholar
Zhu, Y. (2020). A graphon approach to limiting spectral distributions of Wigner-type matrices. Random Structures Algorithms 56, 251–279.CrossRefGoogle Scholar