Probabilistic tensor analysis with akaike and Bayesian information criteria

Dacheng Tao, Jimeng Sun, Xindong Wu, Xuelong Li, Jialie Shen, Stephen J. Maybank, Christos Faloutsos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

From data mining to computer vision, from visual surveillance to biometrics research, from biomedical imaging to bioinformatics, and from multimedia retrieval to information management, a large amount of data are naturally represented by multidimensional arrays, i.e., tensors. However, conventional probabilistic graphical models with probabilistic inference only model data in vector format, although they are very important in many statistical problems, e.g., model selection. Is it possible to construct multilinear probabilistic graphical models for tensor format data to conduct probabilistic inference, e.g., model selection? This paper provides a positive answer based on the proposed decoupled probabilistic model by developing the probabilistic tensor analysis (PTA), which selects suitable model for tensor format data modeling based on Akaike information criterion (AIC) and Bayesian information criterion (BIC). Empirical studies demonstrate that PTA associated with AIC and BIC selects correct number of models.

Original languageEnglish (US)
Title of host publicationNeural Information Processing - 14th International Conference, ICONIP 2007, Revised Selected Papers
Pages791-801
Number of pages11
EditionPART 1
DOIs
StatePublished - Oct 27 2008
Externally publishedYes
Event14th International Conference on Neural Information Processing, ICONIP 2007 - Kitakyushu, Japan
Duration: Nov 13 2007Nov 16 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume4984 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Conference on Neural Information Processing, ICONIP 2007
CountryJapan
CityKitakyushu
Period11/13/0711/16/07

Keywords

  • Akaike information criterion
  • Bayesian information criterion
  • Probabilistic inference
  • Probabilistic principal component analysis
  • Tensor

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Probabilistic tensor analysis with akaike and Bayesian information criteria'. Together they form a unique fingerprint.

  • Cite this

    Tao, D., Sun, J., Wu, X., Li, X., Shen, J., Maybank, S. J., & Faloutsos, C. (2008). Probabilistic tensor analysis with akaike and Bayesian information criteria. In Neural Information Processing - 14th International Conference, ICONIP 2007, Revised Selected Papers (PART 1 ed., pp. 791-801). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4984 LNCS, No. PART 1). https://doi.org/10.1007/978-3-540-69158-7_82