Multivariate Time-Series Classification Using the Hidden-Unit Logistic Model

Wenjie Pei, Hamdi Dibeklioglu, David M.J. Tax, Laurens van der Maaten

Research output: Contribution to journalArticleScientificpeer-review

43 Citations (Scopus)
65 Downloads (Pure)


We present a new model for multivariate time-series classification, called the hidden-unit logistic model (HULM), that uses binary stochastic hidden units to model latent structure in the data. The hidden units are connected in a chain structure that models temporal dependencies in the data. Compared with the prior models for time-series classification such as the hidden
conditional random field, our model can model very complex decision boundaries, because the number of latent states grows exponentially with the number of hidden units. We demonstrate the strong performance of our model in experiments on a variety of (computer vision) tasks, including handwritten character recognition, speech recognition, facial expression, and action recognition. We also present a state-of-the-art system for facial
action unit detection based on the HULM.
Original languageEnglish
Pages (from-to)920-931
Number of pages12
JournalIEEE Transactions on Neural Networks and Learning Systems
Issue number4
Publication statusPublished - 2018

Bibliographical note

Accepted Author Manuscript


  • Hidden unit
  • latent structure modeling
  • temporal dependences modeling
  • time-series classification


Dive into the research topics of 'Multivariate Time-Series Classification Using the Hidden-Unit Logistic Model'. Together they form a unique fingerprint.

Cite this