A tutorial on variational Bayes for latent linear stochastic time-series models (2014)

Abstract

Variational Bayesian methods for the identification of latent stochastic time-series models comprising both observed and unobserved random variables have recently gained momentum in machine learning, theoretical neuroscience, and neuroimaging methods development. Despite their established use as a computationally efficient alternative to sampling-based methods, their practical application in mathematical psychology has so far been limited. In this tutorial we attempt to provide an introductory overview of the theoretical underpinnings that the variational Bayesian approach to latent stochastic time-series models rests on by discussing its application in the linear case. ?? 2014.

Bibliographic entry

Ostwald, D., Kirilina, E., Starke, L., & Blankenburg, F. (2014). A tutorial on variational Bayes for latent linear stochastic time-series models. Journal of Mathematical Psychology, 60, 1-19. doi:10.1016/j.jmp.2014.04.003 (Full text)

Miscellaneous

Publication year 2014
Document type: Article
Publication status: Published
External URL: http://dx.doi.org/10.1016/j.jmp.2014.04.003 View
Categories:
Keywords:

Edit | Publications overview