Learning Essential Graph Markov Models From Data

@inproceedings{Castelo2002LearningEG,
  title={Learning Essential Graph Markov Models From Data},
  author={Robert Castelo and Michael D. Perlman},
  booktitle={Probabilistic Graphical Models},
  year={2002}
}
In a model selection procedure where many models are to be compared, computational efficiency is critical. For acyclic digraph (ADG) Markov models (aka DAG models or Bayesian networks), each ADG Markov equivalence class can be represented by a unique chain graph, called an essential graph (EG). This parsimonious representation might be used to facilitate selection among ADG models. Because EGs combine features of decomposable graphs and ADGs, a scoring metric can be developed for EGs with… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 20 references

A characterization of Markov equivalence classes for acyclic digraphs

S. Andersson, D. Madigan, M. Perlman
Annals of Statistics, 25:505–541. • 1997
View 5 Excerpts
Highly Influenced

Hyper-Markov laws in the statistical analysis of decomposable graphical models

P. Dawid, S. Lauritzen
Annals of Statistics, 21(3):1272– 1317. • 1993
View 16 Excerpts
Highly Influenced

Optimal Structure Identification With Greedy Search

Journal of Machine Learning Research • 2002
View 6 Excerpts
Highly Influenced

Probabilistic Networks and Expert Systems

Information Science and Statistics • 1999
View 2 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…