Tightening Bounds for Variational Inference by Revisiting Perturbation Theory.
Robert BamlerCheng ZhangManfred OpperStephan MandtPublished in: CoRR (2019)
Keyphrases
- variational inference
- perturbation theory
- bayesian inference
- posterior distribution
- topic models
- probabilistic model
- probabilistic graphical models
- gaussian process
- mixture model
- variational methods
- latent dirichlet allocation
- closed form
- upper bound
- lower bound
- exponential family
- exact inference
- maximal cliques
- graphical models
- approximate inference
- bayesian framework
- worst case
- latent variables
- generative model
- maximum likelihood
- level set
- density estimation
- posterior probability
- machine learning
- conditional random fields
- knowledge representation
- particle filter
- factor graphs
- pairwise
- expectation maximization
- model selection