Predictive Distribution of the Dirichlet Mixture Model by Local Variational Inference
2014 (English)In: Journal of Signal Processing Systems, ISSN 1939-8018, Vol. 74, no 3, 359-374 p.Article in journal (Refereed) Published
In Bayesian analysis of a statistical model, the predictive distribution is obtained by marginalizing over the parameters with their posterior distributions. Compared to the frequently used point estimate plug-in method, the predictive distribution leads to a more reliable result in calculating the predictive likelihood of the new upcoming data, especially when the amount of training data is small. The Bayesian estimation of a Dirichlet mixture model (DMM) is, in general, not analytically tractable. In our previous work, we have proposed a global variational inference-based method for approximately calculating the posterior distributions of the parameters in the DMM analytically. In this paper, we extend our previous study for the DMM and propose an algorithm to calculate the predictive distribution of the DMM with the local variational inference (LVI) method. The true predictive distribution of the DMM is analytically intractable. By considering the concave property of the multivariate inverse beta function, we introduce an upper-bound to the true predictive distribution. As the global minimum of this upper-bound exists, the problem is reduced to seek an approximation to the true predictive distribution. The approximated predictive distribution obtained by minimizing the upper-bound is analytically tractable, facilitating the computation of the predictive likelihood. With synthesized data and real data evaluations, the good performance of the proposed LVI based method is demonstrated by comparing with some conventionally used methods.
Place, publisher, year, edition, pages
2014. Vol. 74, no 3, 359-374 p.
Predictive distribution, Dirichlet mixture model, Bayesian inference, Local variational inference
IdentifiersURN: urn:nbn:se:kth:diva-144557DOI: 10.1007/s11265-013-0769-8ISI: 000333206500007ScopusID: 2-s2.0-84897460170OAI: oai:DiVA.org:kth-144557DiVA: diva2:714136
QC 201404252014-04-252014-04-242014-04-25Bibliographically approved