High dimensional changepoint detection with a dynamic graphical lasso
- 1 May 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 2684-2688
Abstract
The use of sparsity to encourage parsimony in graphical models continues to attract much attention at the interface between multivariate Signal Processing and Statistics. We propose and investigate two approaches for the detection of changepoints in the correlation structure of evolving Gaussian graphical models. Both approaches employ two-stages; first estimating the dynamic graphical structure through regularising the precision matrix, before changepoints are selected via a group fused lasso. Experiments on simulated data illustrate the efficacy of the two approaches. Furthermore, results on real internet traffic flow data containing a Denial Of Service attack demonstrate that the proposed approaches have potential utility in information forensics and security.Keywords
This publication has 18 references indexed in Scilit:
- The Joint Graphical Lasso for Inverse Covariance Estimation Across Multiple ClassesJournal of the Royal Statistical Society Series B: Statistical Methodology, 2013
- Sparse graphical modeling of piecewise-stationary time series2011
- Graphical modelling of multivariate time seriesProbability Theory and Related Fields, 2011
- Multiple Change-Point Estimation With a Total Variation PenaltyJournal of the American Statistical Association, 2010
- The benefit of group sparsityThe Annals of Statistics, 2010
- Time varying undirected graphsMachine Learning, 2010
- Recovering time-varying networks of dependencies in social and biological studiesProceedings of the National Academy of Sciences, 2009
- Sparse inverse covariance estimation with the graphical lassoBiostatistics, 2007
- Sparsity and Smoothness Via the Fused LassoJournal of the Royal Statistical Society Series B: Statistical Methodology, 2004
- Graphical ModelsStatistical Science, 2004