A Distributed Laplacian Solver and its Applications to Electrical Flow and Random Spanning Tree Computation

We use queueing networks to present a new approach to solving Laplacian systems. This marks a significant departure from the existing techniques, mostly based on graph-theoretic constructions and sampling. Our distributed solver works for a large and important class of Laplacian systems that we call "one-sink" Laplacian systems. Specifically, our solver can produce solutions for systems of the form $Lx = b$ where exactly one of the coordinates of $b$ is negative. Our solver is a distributed algorithm that takes $\widetilde{O}(t_{hit} d_{\max})$ time (where $\widetilde{O}$ hides $\text{poly}\log n$ factors) to produce an approximate solution where $t_{hit}$ is the worst-case hitting time of the random walk on the graph, which is $\Theta(n)$ for a large set of important graphs, and $d_{\max}$ is the generalized maximum degree of the graph. The class of one-sink Laplacians includes the important voltage computation problem and allows us to compute the effective resistance between nodes in a distributed setting. As a result, our Laplacian solver can be used to adapt the approach by Kelner and M\k{a}dry (2009) to give the first distributed algorithm to compute approximate random spanning trees efficiently.

[1]  Wojciech Szpankowski,et al.  Stability of token passing rings , 1992, Queueing Syst. Theory Appl..

[2]  Richard Peng,et al.  An efficient parallel solver for SDD linear systems , 2013, STOC.

[3]  Pooja Vyavahare,et al.  Decentralized Random Walk-Based Data Collection in Networks , 2017, ArXiv.

[4]  Navin Goyal,et al.  Expanders via random spanning trees , 2008, SODA.

[5]  Fan Chung Graham,et al.  Solving Local Linear Systems with Boundary Conditions Using Heat Kernel Pagerank , 2015, Internet Math..

[6]  John R. Gilbert,et al.  An Empirical Comparison of Graph Laplacian Solvers , 2016, ALENEX.

[7]  Gary L. Miller,et al.  Combinatorial preconditioners and multilevel solvers for problems in computer vision and image processing , 2011, Comput. Vis. Image Underst..

[8]  Haitham Bou-Ammar,et al.  Distributed lifelong reinforcement learning with sub-linear regret , 2017, 2017 IEEE 56th Annual Conference on Decision and Control (CDC).

[9]  Richard L. Tweedie,et al.  Geometric Convergence Rates for Stochastically Ordered Markov Chains , 1996, Math. Oper. Res..

[10]  David Aldous,et al.  The Random Walk Construction of Uniform Spanning Trees and Uniform Labelled Trees , 1990, SIAM J. Discret. Math..

[11]  Nikhil Srivastava,et al.  Graph Sparsification by Effective Resistances , 2011, SIAM J. Comput..

[12]  Alexandr Andoni,et al.  On Solving Linear Systems in Sublinear Time , 2018, ITCS.

[13]  Gary L. Miller,et al.  A Nearly-m log n Time Solver for SDD Linear Systems , 2011, 2011 IEEE 52nd Annual Symposium on Foundations of Computer Science.

[14]  Sekhar Tatikonda,et al.  A new approach to Laplacian solvers and flow problems , 2016, J. Mach. Learn. Res..

[15]  Luca Becchetti,et al.  Pooling or Sampling: Collective Dynamics for Electrical Flow Estimation , 2018, AAMAS.

[16]  Aleksander Madry,et al.  Faster Generation of Random Spanning Trees , 2009, 2009 50th Annual IEEE Symposium on Foundations of Computer Science.

[17]  Richard Peng,et al.  Current Flow Group Closeness Centrality for Complex Networks? , 2018, WWW.

[18]  Artem Napov,et al.  An Efficient Multigrid Method for Graph Laplacian Systems II: Robust Aggregation , 2017, SIAM J. Sci. Comput..

[19]  Jed Brown,et al.  A Parallel Solver for Graph Laplacians , 2018, Proceedings of the Platform for Advanced Scientific Computing Conference.

[20]  Ioannis Koutis,et al.  Simple Parallel and Distributed Algorithms for Spectral Graph Sparsification , 2016, TOPC.

[21]  Noga Alon,et al.  Many random walks are faster than one , 2007, SPAA '08.

[22]  Aaron Schild,et al.  An almost-linear time algorithm for uniform random spanning tree generation , 2017, STOC.

[23]  Aleksander Madry,et al.  Computing Maximum Flow with Augmenting Electrical Flows , 2016, 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS).

[24]  Zeyuan Allen Zhu,et al.  A simple, combinatorial algorithm for solving SDD systems in nearly-linear time , 2013, STOC '13.

[25]  Elizabeth L. Wilmer,et al.  Markov Chains and Mixing Times , 2008 .

[26]  Frank Thomson Leighton,et al.  Multicommodity max-flow min-cut theorems and their use in designing approximation algorithms , 1999, JACM.

[27]  Aleksander Madry,et al.  Fast Generation of Random Spanning Trees and the Effective Resistance Metric , 2015, SODA.

[28]  Achi Brandt,et al.  Lean Algebraic Multigrid (LAMG): Fast Graph Laplacian Linear Solver , 2011, SIAM J. Sci. Comput..

[29]  Bernhard Haeupler,et al.  Near-Optimal BFS-Tree Construction in Radio Networks , 2016, IEEE Communications Letters.

[30]  Gary L. Miller,et al.  Parallel graph decompositions using random shifts , 2013, SPAA.

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

[32]  Michael Wegner,et al.  Is Nearly-linear the Same in Theory and Practice? A Case Study with a Combinatorial Laplacian Solver , 2015, SEA.

[33]  Carl M. Harris,et al.  Fundamentals of queueing theory , 1975 .

[34]  David Peleg,et al.  Distributed Computing: A Locality-Sensitive Approach , 1987 .

[35]  Pooja Vyavahare,et al.  A Stochastic Process on a Network with Connections to Laplacian Systems of Equations , 2017 .

[36]  Nikolaos M. Freris,et al.  Randomized gossip algorithms for solving Laplacian systems , 2015, 2015 European Control Conference (ECC).

[37]  Sushant Sachdeva,et al.  Approximate Gaussian Elimination for Laplacians - Fast, Sparse, and Simple , 2016, 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS).

[38]  Baruch Awerbuch,et al.  Distributed BFS algorithms , 1985, 26th Annual Symposium on Foundations of Computer Science (sfcs 1985).

[39]  Richard Peng,et al.  Approximate Undirected Maximum Flows in O(mpolylog(n)) Time , 2014, SODA.

[40]  Shang-Hua Teng,et al.  Nearly-linear time algorithms for graph partitioning, graph sparsification, and solving linear systems , 2003, STOC '04.

[41]  Devdatt P. Dubhashi,et al.  Balls and bins: A study in negative dependence , 1996, Random Struct. Algorithms.

[42]  Haitham Bou-Ammar,et al.  A Fast Distributed Solver for Symmetric Diagonally Dominant Linear Equations , 2015, ArXiv.

[43]  Gary L. Miller,et al.  A linear work, O(n1/6) time, parallel algorithm for solving planar Laplacians , 2007, SODA '07.

[44]  Andrei Z. Broder,et al.  Generating random spanning trees , 1989, 30th Annual Symposium on Foundations of Computer Science.

[45]  Guy E. Blelloch,et al.  Near linear-work parallel SDD solvers, low-diameter decomposition, and low-stretch subgraphs , 2011, SPAA '11.

[46]  O. Rojo,et al.  Ordering trees and graphs with few cycles by algebraic connectivity , 2014 .

[47]  Shang-Hua Teng,et al.  Electrical flows, laplacian systems, and faster approximation of maximum flow in undirected graphs , 2010, STOC '11.

[48]  Peter G. Doyle,et al.  Random Walks and Electric Networks: REFERENCES , 1987 .

[49]  W. Hoeffding Probability Inequalities for sums of Bounded Random Variables , 1963 .

[50]  John Peebles,et al.  Sampling random spanning trees faster than matrix multiplication , 2016, STOC.

[51]  Rasul Tutunov,et al.  Fully Distributed and Mixed Symmetric Diagonal Dominant Solvers for Large Scale Optimization , 2017 .

[52]  Wendi B. Heinzelman,et al.  Adaptive protocols for information dissemination in wireless sensor networks , 1999, MobiCom.

[53]  Richard J. Lipton,et al.  Random walks, universal traversal sequences, and the complexity of maze problems , 1979, 20th Annual Symposium on Foundations of Computer Science (sfcs 1979).