Message Passing-Based Inference in the Gamma Mixture Model

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

The Gamma mixture model is a flexible probability distribution for representing beliefs about scale variables such as precisions. Inference in the Gamma mixture model for all latent variables is non-trivial as it leads to intractable equations. This paper presents two variants of variational message passing-based inference in a Gamma mixture model. We use moment matching and alternatively expectation-maximization to approximate the posterior distributions. The proposed method supports automated inference in factor graphs for large probabilistic models that contain multiple Gamma mixture models as plug-in factors. The Gamma mixture model has been implemented in a factor graph package and we present experimental results for both synthetic and real-world data sets.
Original languageEnglish
Title of host publication2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)
PublisherInstitute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)978-1-7281-6338-3
ISBN (Print)978-1-6654-1184-4
DOIs
Publication statusPublished - 15 Nov 2021
Event31st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2021 - Virtual, Gold Coast, Australia
Duration: 25 Oct 202128 Oct 2021
Conference number: 31
https://2021.ieeemlsp.org/

Conference

Conference31st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2021
Abbreviated titleMLSP 2021
Country/TerritoryAustralia
CityGold Coast
Period25/10/2128/10/21
Internet address

Keywords

  • Mixture models
  • Machine learning
  • Signal processing
  • Probabilistic logic
  • Mathematical models
  • Probability distribution

Fingerprint

Dive into the research topics of 'Message Passing-Based Inference in the Gamma Mixture Model'. Together they form a unique fingerprint.

Cite this