A Collapsed Variational Bayesian Inference Algorithm for Latent Dirichlet Allocation

Abstract

Latent Dirichlet allocation (LDA) is a Bayesian network that has recently gained much popularity in applications ranging from document modeling to computer vision. Due to the large scale nature of these applications, current inference procedures like variational Bayes and Gibbs sampling have been found lacking. In this paper we propose the collapsed variational Bayesian inference algorithm for LDA, and show that it is computationally efficient, easy to implement and significantly more accurate than standard variational Bayesian inference for LDA.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 07, 2007
Accession Number
ADA629956

Entities

People

  • David Newman
  • Max Welling
  • Yee W. Teh

Organizations

  • University of California, Irvine

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Bayesian Inference
  • Bayesian Networks
  • Computational Science
  • Computer Science
  • Computers
  • Data Science
  • Equations
  • Free Energy
  • Information Science
  • Models
  • Monte Carlo Method
  • Probability
  • Random Variables
  • Sampling
  • Standards
  • Test Sets

Fields of Study

  • Computer science

Readers

  • Psychometric Testing or Psychological Assessment.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Information Retrieval
  • AI & ML - Machine Learning Algorithms