Graph-Based Ascent Algorithms for Function Maximization

We study the problem of finding the maximum of a function defined on the nodes of a connected graph. The goal is to identify a node where the function obtains its maximum. We focus on local iterative algorithms, which traverse the nodes of the graph along a path, and the next iterate is chosen from the neighbors of the current iterate with probability distribution determined by the function values at the current iterate and its neighbors. We study two algorithms corresponding to a Metropolis-Hastings random walk with different transition kernels: (i) The first algorithm is an exponentially weighted random walk governed by a parameter gamma. (ii) The second algorithm is defined with respect to the graph Laplacian and a smoothness parameter k. We derive convergence rates for the two algorithms in terms of total variation distance and hitting times. We also provide simulations showing the relative convergence rates of our algorithms in comparison to an unbiased random walk, as a function of the smoothness of the graph function. Our algorithms may be categorized as a new class of “descent-based” methods for function maximization on the nodes of a graph.

[1]  Alan M. Frieze,et al.  Looking for vertex number one , 2014, ArXiv.

[2]  Alexander J. Smola,et al.  Kernels and Regularization on Graphs , 2003, COLT.

[3]  Pierre Vandergheynst,et al.  Random sampling of bandlimited signals on graphs , 2015, NIPS 2015.

[4]  Pascal Frossard,et al.  The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains , 2012, IEEE Signal Processing Magazine.

[5]  Ulrike von Luxburg,et al.  A tutorial on spectral clustering , 2007, Stat. Comput..

[6]  Pierre Vandergheynst,et al.  Compressive Spectral Clustering , 2016, ICML.

[7]  K. Schittkowski,et al.  NONLINEAR PROGRAMMING , 2022 .

[8]  Yee Whye Teh,et al.  Bayesian Learning via Stochastic Gradient Langevin Dynamics , 2011, ICML.

[9]  Antonio Ortega,et al.  Submitted to Ieee Transactions on Signal Processing 1 Efficient Sampling Set Selection for Bandlimited Graph Signals Using Graph Spectral Proxies , 2022 .

[10]  Michael Kearns,et al.  Local Algorithms for Finding Interesting Individuals in Large Networks , 2010, ICS.

[11]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[12]  Sanjeev Khanna,et al.  The Power of Local Information in Social Networks , 2012, WINE.

[13]  Y. Nesterov.  A method for solving the convex programming problem with convergence rate O(1/k^2) , 1983 .

[14]  John D. Lafferty,et al.  Diffusion Kernels on Graphs and Other Discrete Input Spaces , 2002, ICML.

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

[16]  Alexander J. Smola,et al.  Trend Filtering on Graphs , 2014, J. Mach. Learn. Res..

[17]  Yuchen Zhang,et al.  A Hitting Time Analysis of Stochastic Gradient Langevin Dynamics , 2017, COLT.

[18]  S. Yau,et al.  Ricci curvature of graphs , 2011 .

[19]  Ali Esmaili,et al.  Probability and Random Processes , 2005, Technometrics.

[20]  Yurii Nesterov,et al.  Introductory Lectures on Convex Optimization - A Basic Course , 2014, Applied Optimization.

[21]  Douglas D. Heckathorn,et al.  Respondent-driven sampling : A new approach to the study of hidden populations , 1997 .

[22]  Yicheng Zhang,et al.  Identifying influential nodes in complex networks , 2012 .

[23]  Abhijit Banerjee,et al.  Gossip: Identifying Central Individuals in a Social Network , 2014, ArXiv.

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

[25]  Hiroshi Hirai,et al.  Discrete Convex Functions on Graphs and Their Algorithmic Applications , 2017, 1706.09106.

[26]  Matus Telgarsky,et al.  Non-convex learning via Stochastic Gradient Langevin Dynamics: a nonasymptotic analysis , 2017, COLT.

[27]  A. Dalalyan.  Theoretical guarantees for approximate sampling from smooth and log‐concave densities , 2014, 1412.7392.