Random Walks on Dynamic Graphs: Mixing Times, HittingTimes, and Return Probabilities

We establish and generalise several bounds for various random walk quantities including the mixing time and the maximum hitting time. Unlike previous analyses, our derivations are based on rather intuitive notions of local expansion properties which allows us to capture the progress the random walk makes through $t$-step probabilities. We apply our framework to dynamically changing graphs, where the set of vertices is fixed while the set of edges changes in each round. For random walks on dynamic connected graphs for which the stationary distribution does not change over time, we show that their behaviour is in a certain sense similar to static graphs. For example, we show that the mixing and hitting times of any sequence of $d$-regular connected graphs is $O(n^2)$, generalising a well-known result for static graphs. We also provide refined bounds depending on the isoperimetric dimension of the graph, matching again known results for static graphs. Finally, we investigate properties of random walks on dynamic graphs that are not always connected: we relate their convergence to stationarity to the spectral properties of an average of transition matrices and provide some examples that demonstrate strong discrepancies between static and dynamic graphs.

[1]  Paul G. Spirakis,et al.  Cover Time in Edge-Uniform Stochastically-Evolving Graphs , 2017, SSS.

[2]  Luca Trevisan,et al.  Information spreading in dynamic graphs , 2011, PODC '12.

[3]  J. A. Fill.  Eigenvalue bounds on convergence to stationarity for nonreversible markov chains , 1991 .

[4]  Luís E. T. Rodrigues,et al.  Random Walks on Evolving Graphs with Recurring Topologies , 2014, DISC.

[5]  V. Climenhaga.  Markov chains and mixing times , 2013 .

[6]  Zhenming Liu,et al.  Information Dissemination via Random Walks in d-Dimensional Space , 2011, SODA.

[7]  Prabhakar Raghavan,et al.  The electrical resistance of a graph captures its commute and cover times , 2005, computational complexity.

[8]  Yuval Peres,et al.  Mobile geometric graphs: detection, coverage and percolation , 2010, Probability Theory and Related Fields.

[9]  Eli Upfal,et al.  Tight bounds on information dissemination in sparse mobile networks , 2011, PODC '11.

[10]  Perla Sousi,et al.  A comparison principle for random walk on dynamical percolation , 2019, The Annals of Probability.

[11]  Uriel Feige,et al.  A Tight Lower Bound on the Cover Time for Random Walks on Graphs , 1995, Random Struct. Algorithms.

[12]  Y. Peres,et al.  Random walks on dynamical percolation: mixing times, mean squared displacement and hitting times , 2013, 1308.6193.

[13]  George Giakkoupis,et al.  Bounds on the Voter Model in Dynamic Networks , 2016, ICALP.

[14]  L. Saloff-Coste,et al.  Merging for inhomogeneous finite Markov chains, part II: Nash and log-Sobolev inequalities , 2011, 1104.1560.

[15]  Andrea E. F. Clementi,et al.  Rumor spreading in random evolving graphs , 2013, Random Struct. Algorithms.

[16]  Paul G. Spirakis,et al.  Cover Time in Edge-Uniform Stochastically-Evolving Graphs , 2018, Algorithms.

[17]  Uriel Feige,et al.  A Tight Upper Bound on the Cover Time for Random Walks on Graphs , 1995, Random Struct. Algorithms.

[18]  Fabian Kuhn,et al.  Dynamic networks: models and algorithms , 2011, SIGA.

[19]  P. Tetali,et al.  Mixing Time Bounds via the Spectral Profile , 2005, math/0505690.

[20]  Perla Sousi,et al.  Cutoff for Random Walk on Dynamical Erd\H{o}s-R\'enyi Graph , 2018 .

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

[22]  Perla Sousi,et al.  Cutoff for random walk on dynamical Erdős–Rényi graph , 2018, Annales de l'Institut Henri Poincaré, Probabilités et Statistiques.

[23]  J. Welch.  Distributed Algorithmic Foundations of Dynamic Networks , 2016 .

[24]  Shayan Oveis Gharan.  Recent Advances in Approximation Algorithms Spring 2015 Lecture 4-5 : Effective Resistance and Simple Random Walks , 2015 .

[25]  László Lovász,et al.  Faster mixing via average conductance , 1999, STOC '99.

[26]  Nancy A. Lynch,et al.  Distributed computation in dynamic networks , 2010, STOC '10.

[27]  Thomas Sauerwald,et al.  On Coalescence Time in Graphs: When Is Coalescing as Fast as Meeting? , 2016, SODA.

[28]  Chen Avin,et al.  Cover time and mixing time of random walks on dynamic graphs , 2018, Random Struct. Algorithms.

[29]  Perla Sousi,et al.  Random walk on dynamical percolation , 2019 .

[30]  Prabhakar Raghavan,et al.  The electrical resistance of a graph captures its commute and cover times , 1989, STOC '89.

[31]  Paul G. Spirakis,et al.  Elements of the theory of dynamic networks , 2018, Commun. ACM.

[32]  Andrea E. F. Clementi,et al.  Flooding Time of Edge-Markovian Evolving Graphs , 2010, SIAM J. Discret. Math..

[33]  Bruce M. Maggs,et al.  Tight analyses of two local load balancing algorithms , 1995, STOC '95.

[34]  Gady Kozma,et al.  A Resistance Bound Via An Isoperimetric Inequality , 2005, Comb..

[35]  Robert Elsässer,et al.  Load balancing in dynamic networks , 2004, 7th International Symposium on Parallel Architectures, Algorithms and Networks, 2004. Proceedings..

[36]  Mark Jerrum,et al.  Approximate Counting, Uniform Generation and Rapidly Mixing Markov Chains , 1987, International Workshop on Graph-Theoretic Concepts in Computer Science.

[37]  David J. Aldous,et al.  Lower bounds for covering times for reversible Markov chains and random walks on graphs , 1989 .

[38]  George Giakkoupis,et al.  Randomized Rumor Spreading in Dynamic Graphs , 2014, ICALP.