Variational Bayesian partially observed non-negative tensor factorization

Jesper L. Hinrich, Soren F.V. Nielsen, Kristoffer H. Madsen, Morten Morup

5 Citations (Scopus)

Abstract

Non-negative matrix and tensor factorization (NMF/NTF) have become important tools for extracting part based representations in data. It is however unclear when an NMF or NTF approach is most suited for data and how reliably the models predict when trained on partially observed data. We presently extend a recently proposed variational Bayesian NMF (VB-NMF) to non-negative tensor factorization (VB-NTF) for partially observed data. This admits bi- and multi-linear structure quantification considering both model prediction and evidence. We evaluate the developed VB-NTF on synthetic and a real dataset of gene expression in the human brain and contrast the performance to VB-NMF and conventional NMF/NTF. We find that the gene expressions are better accounted for by VB-NMF than VB-NTF and that VB-NMF/VB-NTF more robustly handle partially observed data than conventional NMF/NTF. In particular, probabilistic modeling is beneficial when large amounts of data is missing and/or the model order over-specified.

Original languageEnglish
Title of host publication2018 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018 - Proceedings
Volume2018-September
PublisherIEEE Computer Society Press
Publication date2018
Pages1-6
Article number8516924
ISBN (Electronic)9781538654774
DOIs
Publication statusPublished - 2018
Event28th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018 - Aalborg, Denmark
Duration: 17 Sept 201820 Sept 2018

Conference

Conference28th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018
Country/TerritoryDenmark
CityAalborg
Period17/09/201820/09/2018
SponsorIEEE Signal Processing Society

Keywords

  • Human brain microarray data
  • Missing data
  • Non-negative tensor factorization
  • Probabilistic modeling

Fingerprint

Dive into the research topics of 'Variational Bayesian partially observed non-negative tensor factorization'. Together they form a unique fingerprint.

Cite this