Comparison Of Local And Global Undirected Graphical Models

Zhemin Zhu, Djoerd Hiemstra, Peter M.G. Apers, Andreas Wombacher

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

2 Citations (Scopus)
9 Downloads (Pure)

Abstract

CRFs are discriminative undirected models which are globally normalized. Global normalization preserves CRFs from the label bias problem which most local models suffer from. Recently proposed co-occurrence rate networks (CRNs) are also discriminative undirected models. In contrast to CRFs, CRNs are locally normalized. It was established that CRNs are immune to the label bias problem even they are local models. In this paper, we further compare ECRNs (using fully empirical relative frequencies, not by support vector Regression) and CRFs. The connection between Co-occurrence Rate, which is the exponential function of pointwise mutual information, and Copulas is built in continuous case. Also they are further evaluated statistically by experiments.
Original languageEnglish
Title of host publicationESANN 2014
Subtitle of host publication22nd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning : Bruges, Belgium, April 23-24-25, 2014 : proceedings
EditorsMichel Verleysen
Place of PublicationLouvain-la-Neuve, Begium
Publisheri6doc.com
Pages479-484
Number of pages6
ISBN (Print)978-2-87419-095-7
Publication statusPublished - Apr 2014

    Fingerprint

Keywords

  • EWI-24923
  • IR-91898
  • METIS-305953

Cite this

Zhu, Z., Hiemstra, D., Apers, P. M. G., & Wombacher, A. (2014). Comparison Of Local And Global Undirected Graphical Models. In M. Verleysen (Ed.), ESANN 2014: 22nd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning : Bruges, Belgium, April 23-24-25, 2014 : proceedings (pp. 479-484). Louvain-la-Neuve, Begium: i6doc.com.