A low discrepancy sequence on graphs

Many applications such as election forecasting, environmental monitoring, health policy, and graph based machine learning require taking expectation of functions defined on the vertices of a graph. We describe a construction of a sampling scheme analogous to the so called Leja points in complex potential theory that can be proved to give low discrepancy estimates for the approximation of the expected value by the impirical expected value based on these points. In contrast to classical potential theory where the kernel is fixed and the equilibrium distribution depends upon the kernel, we fix a probability distribution and construct a kernel (which represents the graph structure) for which the equilibrium distribution is the given probability distribution. Our estimates do not depend upon the size of the graph.

[1]  Daniel Gayo-Avello,et al.  A Meta-Analysis of State-of-the-Art Electoral Prediction From Twitter Data , 2012, ArXiv.

[2]  Bent Fuglede,et al.  On the theory of potentials in locally compact spaces , 1960 .

[3]  James K. Harter,et al.  The well-being 5: development and validation of a diagnostic instrument to improve population well-being. , 2014, Population health management.

[4]  N. S. Landkof.  Foundations of Modern Potential Theory , 1972 .

[5]  Gautam Das,et al.  Walk, Not Wait: Faster Sampling Over Online Social Networks , 2014, Proc. VLDB Endow..

[6]  Kristina Lerman,et al.  The Role of Social Media in the Discussion of Controversial Topics , 2013, 2013 International Conference on Social Computing.

[7]  Ronald R. Coifman,et al.  Doubly-Stochastic Normalization of the Gaussian Kernel is Robust to Heteroskedastic Noise , 2021, SIAM J. Math. Data Sci..

[8]  H. Mhaskar.  Weighted polynomials, radial basis functions and potentials on locally compact spaces , 1990 .

[9]  Toshihisa Tanaka,et al.  Eigendecomposition-Free Sampling Set Selection for Graph Signals , 2018, IEEE Transactions on Signal Processing.

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

[11]  Mario Götz,et al.  On the Distribution of Leja-Górski Points , 2001 .

[12]  Hrushikesh N. Mhaskar.  Dimension independent bounds for general shallow networks , 2020, Neural Networks.

[13]  Stefano De Marchi,et al.  On Leja sequences: some results and applications , 2004, Appl. Math. Comput..

[14]  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 .

[15]  Andreas Krause,et al.  Efficient Sensor Placement Optimization for Securing Large Water Distribution Networks , 2008 .

[16]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[17]  Kristina Lerman,et al.  Tripartite graph clustering for dynamic sentiment analysis on social media , 2014, SIGMOD Conference.

[18]  I. Pesenson.  Sampling in paley-wiener spaces on combinatorial graphs , 2008, 1111.5896.

[19]  Stefan Steinerberger,et al.  Numerical Integration on Graphs: where to sample and how to weigh , 2020, Math. Comput..

[20]  P. Erdös,et al.  On the Uniformly-Dense Distribution of Certain Sequences of Points , 1940 .

[21]  Louis Brown.  Sequences of Well-Distributed Vertices on Graphs and Spectral Bounds on Optimal Transport , 2020, ArXiv.

[22]  Edward M. Reingold,et al.  Graph drawing by force‐directed placement , 1991, Softw. Pract. Exp..

[23]  R. Brualdi.  The DAD theorem for arbitrary row sums , 1974 .

[24]  Hrushikesh Narhar Mhaskar,et al.  On the tractability of multivariate integration and approximation by neural networks , 2004, J. Complex..

[25]  Hans-Peter Blatt,et al.  Discrepancy of Signed Measures and Polynomial Approximation , 2001 .

[26]  I. Pritsker.  Equidistribution of points via energy , 2011, 1307.6076.

[27]  Aarti Singh,et al.  Active Learning for Graph Neural Networks via Node Feature Propagation , 2019, ArXiv.

[28]  H. Mhaskar,et al.  A general discrepancy theorem , 1993 .

[29]  Jelena Kovacevic,et al.  Discrete Signal Processing on Graphs: Sampling Theory , 2015, IEEE Transactions on Signal Processing.

[30]  Duncan J. Watts,et al.  Collective dynamics of ‘small-world’ networks , 1998, Nature.

[31]  Stefan Steinerberger,et al.  Generalized designs on graphs: Sampling, spectra, symmetries , 2018, J. Graph Theory.

[32]  Lauwerens Kuipers,et al.  Uniform distribution of sequences , 1974 .

[33]  Stefan Steinerberger,et al.  Quadrature Points via Heat Kernel Repulsion , 2018, Constructive Approximation.

[34]  Philip A. Knight,et al.  The Sinkhorn-Knopp Algorithm: Convergence and Applications , 2008, SIAM J. Matrix Anal. Appl..

[35]  Ronald R. Coifman,et al.  Multiscale data sampling and function extension , 2013 .

[36]  U. Feige,et al.  Spectral Graph Theory , 2015 .

[37]  On the distribution of simple zeros of polynomials , 1992 .

[38]  Fred J. Hickernell,et al.  Randomized Halton sequences , 2000 .

[39]  Raaz Dwivedi,et al.  The power of online thinning in reducing discrepancy , 2016, Probability Theory and Related Fields.

[40]  Lise Getoor,et al.  Collective Classification in Network Data , 2008, AI Mag..

[41]  F. Pillichshammer,et al.  Digital Nets and Sequences: Discrepancy Theory and Quasi-Monte Carlo Integration , 2010 .

[42]  Yonina C. Eldar,et al.  Sampling Signals on Graphs: From Theory to Applications , 2020, IEEE Signal Processing Magazine.

[43]  Richard Sinkhorn,et al.  Concerning nonnegative matrices and doubly stochastic matrices , 1967 .

[44]  Rajeev Motwani,et al.  The PageRank Citation Ranking : Bringing Order to the Web , 1999, WWW 1999.

[45]  F. Leja.  Sur certaines suites liées aux ensembles plans et leur application à la représentation conforme , 1957 .

[46]  M. Ohtsuka.  On potentials in locally compact spaces. , 1961 .

[47]  Ronald R. Coifman,et al.  Manifold learning with bi-stochastic kernels , 2017, IMA Journal of Applied Mathematics.

[48]  Athanasios V. Vasilakos,et al.  Albatross sampling: robust and effective hybrid vertex sampling for social graphs , 2011, HotPlanet '11.

[49]  H. Krumholz,et al.  People mover's distance: Class level geometry using fast pairwise data adaptive transportation costs , 2017, Applied and Computational Harmonic Analysis.