Sequential change-point detection in high-dimensional Gaussian graphical models

High dimensional piecewise stationary graphical models represent a versatile class for modelling time varying networks arising in diverse application areas, including biology, economics, and social sciences. There has been recent work in offline detection and estimation of regime changes in the topology of sparse graphical models. However, the online setting remains largely unexplored, despite its high relevance to applications in sensor networks and other engineering monitoring systems, as well as financial markets. To that end, this work introduces a novel scalable online algorithm for detecting an unknown number of abrupt changes in the inverse covariance matrix of sparse Gaussian graphical models with small delay. The proposed algorithm is based upon monitoring the conditional log-likelihood of all nodes in the network and can be extended to a large class of continuous and discrete graphical models. We also investigate asymptotic properties of our procedure under certain mild regularity conditions on the graph size, sparsity level, number of samples, and pre- and post-changes in the topology of the network. Numerical works on both synthetic and real data illustrate the good performance of the proposed methodology both in terms of computational and statistical efficiency across numerous experimental settings.

[1]  Erwin Bolthausen,et al.  On the Central Limit Theorem for Stationary Mixing Random Fields , 1982 .

[2]  George Michailidis,et al.  Change point estimation in high dimensional Markov random‐field models , 2014, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[3]  T. Cai,et al.  Two-Sample Covariance Matrix Testing and Support Recovery in High-Dimensional and Sparse Settings , 2013 .

[4]  XuanLong Nguyen,et al.  Optimal change point detection in Gaussian processes , 2015, ArXiv.

[5]  T. Cai,et al.  A Constrained ℓ1 Minimization Approach to Sparse Precision Matrix Estimation , 2011, 1102.2233.

[6]  L. Baxter Random Fields on a Network: Modeling, Statistics, and Applications , 1996 .

[7]  George Michailidis,et al.  Adaptive statistical detection of false data injection attacks in smart grids , 2016, 2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[8]  D. Bessler,et al.  The Structure of Interdependence in International Stock Markets , 2003 .

[9]  I JordanMichael,et al.  Graphical Models, Exponential Families, and Variational Inference , 2008 .

[10]  Michèle Basseville,et al.  Detection of abrupt changes: theory and application , 1993 .

[11]  Lin Xiao,et al.  Dual Averaging Methods for Regularized Stochastic Learning and Online Optimization , 2009, J. Mach. Learn. Res..

[12]  Gábor Lugosi,et al.  Concentration Inequalities - A Nonasymptotic Theory of Independence , 2013, Concentration Inequalities.

[13]  A. Gibberd,et al.  Consistent multiple changepoint estimation with fused Gaussian graphical models , 2020 .

[14]  Steven E. Pav Moments of the log non-central chi-square distribution , 2015, 1503.06266.

[15]  Sheldon M. Ross,et al.  Introduction to probability models , 1975 .

[16]  秀俊 松井,et al.  Statistics for High-Dimensional Data: Methods, Theory and Applications , 2014 .

[17]  Michael A. West,et al.  Covariance decomposition in undirected Gaussian graphical models , 2005 .

[18]  N. Zhang,et al.  Graph-based change-point detection , 2012, 1209.1625.

[19]  A. Aue,et al.  Break detection in the covariance structure of multivariate time series models , 2009, 0911.3796.

[20]  Pradeep Ravikumar,et al.  QUIC: quadratic approximation for sparse inverse covariance estimation , 2014, J. Mach. Learn. Res..

[21]  Jukka-Pekka Onnela,et al.  Change Point Detection in Correlation Networks , 2014, Scientific Reports.

[22]  Gustavo Deco,et al.  Can sliding-window correlations reveal dynamic functional connectivity in resting-state fMRI? , 2016, NeuroImage.

[23]  P. Perron,et al.  Estimating and testing linear models with multiple structural changes , 1995 .

[24]  C. Stein A bound for the error in the normal approximation to the distribution of a sum of dependent random variables , 1972 .

[25]  Alex J. Gibberd,et al.  Multiple Changepoint Estimation in High-Dimensional Gaussian Graphical Models , 2017, 1712.05786.

[26]  P. Perron,et al.  Computation and Analysis of Multiple Structural-Change Models , 1998 .

[27]  Yves Atchadé,et al.  Change-Point Computation for Large Graphical Models: A Scalable Algorithm for Gaussian Graphical Models with Change-Points , 2017, J. Mach. Learn. Res..

[28]  Le Song,et al.  Estimating time-varying networks , 2008, ISMB 2008.

[29]  Xindong Wu,et al.  Data mining with big data , 2014, IEEE Transactions on Knowledge and Data Engineering.

[30]  Mladen Kolar,et al.  Estimating networks with jumps. , 2010, Electronic journal of statistics.

[31]  L. Horváth,et al.  Extensions of some classical methods in change point analysis , 2014 .

[32]  Ravi S. Menon,et al.  Resting‐state networks show dynamic functional connectivity in awake humans and anesthetized macaques , 2013, Human brain mapping.

[33]  Jun Yu Li,et al.  Two Sample Tests for High Dimensional Covariance Matrices , 2012, 1206.0917.