Time Varying Undirected Graphs

Simple item page
dc.creatorZhou, Shuheng
dc.creatorLafferty, John
dc.creatorWasserman, Larry
dc.date2008-02-20
dc.date2008-04-29
dc.date.accessioned2026-07-07T09:35:26Z
dc.date.available2026-07-07T09:35:26Z
dc.descriptionUndirected graphs are often used to describe high dimensional distributions. Under sparsity conditions, the graph can be estimated using $\ell_1$ penalization methods. However, current methods assume that the data are independent and identically distributed. If the distribution, and hence the graph, evolves over time then the data are not longer identically distributed. In this paper, we show how to estimate the sequence of graphs for non-identically distributed data, where the distribution evolves over time.
dc.description12 pages, 3 figures, to appear in COLT 2008
dc.identifierhttps://arxiv.org/abs/0802.2758
dc.identifierhttp://arxiv.org/abs/0802.2758
dc.identifierThe 21st Annual Conference on Learning Theory (COLT 2008), Helsinki, Finland
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/159832
dc.subjectMachine Learning
dc.subjectStatistics Theory
dc.titleTime Varying Undirected Graphs
dc.typetext

Files

Collections

arXiv