Entropy-based Variational Learning of Finite Inverted Beta-Liouville Mixture Model

Authors

  • Mohammad Sadegh Ahmadzadeh Concordia University
  • Narges Manouchehri Concordia University https://orcid.org/0000-0002-3011-5162
  • Hafsa Ennajari Concordia University
  • Nizar Bouguila, Professor Concordia University
  • Manar Amayri Grenoble Institute of Technology
  • Wentao Fan Huaqiao University

DOI:

https://doi.org/10.32473/flairs.v34i1.128379

Keywords:

Mixture models, Inverted Beta-Liouville distribution, Entropy-based variational inference

Abstract

Mixture models are a common unsupervised learning technique that have been widely used to statistically approximate and analyse heterogenous data. In this paper, an effective mixture model-based approach for positive vectors clustering and modeling is proposed. Our mixture model is based on the inverted Beta-Liouville (IBL) distribution. To deploy the proposed model, we introduce an entropy-based variational inference algorithm. The performance of the proposed model is evaluated on two real-world applications, namely, human activity recognition and image categorization.

Downloads

Published

2021-04-18

How to Cite

Ahmadzadeh, M. S., Manouchehri, N., Ennajari, H., Bouguila, N., Amayri, M., & Fan, W. (2021). Entropy-based Variational Learning of Finite Inverted Beta-Liouville Mixture Model. The International FLAIRS Conference Proceedings, 34. https://doi.org/10.32473/flairs.v34i1.128379

Issue

Section

Special Track: Uncertain Reasoning