Undirected graphs are often used to describe high dimensional distributions. Under sparsity conditions, the graph can be estimated using ℓ1 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.
@article{arxiv.0802.2758,
title = {Time Varying Undirected Graphs},
author = {Shuheng Zhou and John Lafferty and Larry Wasserman},
journal= {arXiv preprint arXiv:0802.2758},
year = {2008}
}