Average Whenever You Meet: Opportunistic Protocols for Community Detection

Consider the following asynchronous, opportunistic communication model over a graph G: in each round, one edge is activated uniformly and independently at random and (only) its two endpoints can exchange messages and perform local computations. Under this model, we study the following random process: The first time a vertex is an endpoint of an active edge, it chooses a random number, say +/- 1 with probability 1/2; then, in each round, the two endpoints of the currently active edge update their values to their average. We provide a rigorous analysis of the above process showing that, if G exhibits a two-community structure (for example, two expanders connected by a sparse cut), the values held by the nodes will collectively reflect the underlying community structure over a suitable phase of the above process. Our analysis requires new concentration bounds on the product of certain random matrices that are technically challenging and possibly of independent interest. We then exploit our analysis to design the first opportunistic protocols that approximately recover community structure using only logarithmic (or polylogarithmic, depending on the sparsity of the cut) work per node.

[1]  Stuart M. Allen,et al.  Decentralised detection of periodic encounter communities in opportunistic networks , 2012, Ad Hoc Networks.

[2]  Minoru Toda,et al.  Springer Series in Solid-State Sciences , 1989 .

[3]  David Eisenstat,et al.  The computational power of population protocols , 2006, Distributed Computing.

[4]  David Kempe,et al.  A decentralized algorithm for spectral analysis , 2004, STOC '04.

[5]  He Sun,et al.  Distributed Graph Clustering and Sparsification , 2017 .

[6]  Elchanan Mossel,et al.  Reconstruction and estimation in the planted partition model , 2012, Probability Theory and Related Fields.

[7]  Amin Coja-Oghlan,et al.  Graph Partitioning via Adaptive Spectral Techniques , 2009, Combinatorics, Probability and Computing.

[8]  Emmanuel Abbe,et al.  Exact Recovery in the Stochastic Block Model , 2014, IEEE Transactions on Information Theory.

[9]  Ioana Dumitriu,et al.  Recovery and Rigidity in a Regular Stochastic Block Model , 2016, SODA.

[10]  Frank McSherry,et al.  Spectral partitioning of random graphs , 2001, Proceedings 2001 IEEE International Conference on Cluster Computing.

[11]  Cristopher Moore,et al.  Asymptotic analysis of the stochastic block model for modular networks and its algorithmic applications , 2011, Physical review. E, Statistical, nonlinear, and soft matter physics.

[12]  Laurent Massoulié,et al.  Community detection thresholds and the weak Ramanujan property , 2013, STOC.

[13]  He Sun,et al.  Distributed Graph Clustering by Load Balancing , 2016, SPAA.

[14]  Stephen P. Boyd,et al.  Randomized gossip algorithms , 2006, IEEE Transactions on Information Theory.

[15]  Luca Trevisan,et al.  Find Your Place: Simple Distributed Algorithms for Community Detection , 2015, SODA.

[16]  Alessandro Panconesi,et al.  Concentration of Measure for the Analysis of Randomized Algorithms , 2009 .

[17]  J. R. Ipsen Products of independent Gaussian random matrices , 2015, 1510.06128.

[18]  Mark Jerrum,et al.  The Metropolis Algorithm for Graph Bisection , 1998, Discret. Appl. Math..

[19]  Charles Bordenave,et al.  A new proof of Friedman's second eigenvalue theorem and its extension to random lifts , 2015, Annales scientifiques de l'École normale supérieure.

[20]  Martin E. Dyer,et al.  The Solution of Some Random NP-Hard Problems in Polynomial Expected Time , 1989, J. Algorithms.

[21]  Elchanan Mossel,et al.  A Proof of the Block Model Threshold Conjecture , 2013, Comb..

[22]  Elchanan Mossel,et al.  Belief propagation, robust reconstruction and optimal recovery of block models , 2013, COLT.