Joint Bayesian Inference of Graphical Structure and Parameters with a Single Generative Flow Network.
Tristan DeleuMizu Nishikawa-ToomeyJithendaraa SubramanianNikolay MalkinLaurent CharlinYoshua BengioPublished in: CoRR (2023)
Keyphrases
- bayesian inference
- conditional likelihood
- graphical structure
- markov networks
- hyperparameters
- probabilistic model
- graphical models
- generative model
- prior information
- random variables
- bayesian belief networks
- parameter estimation
- objective function
- flow network
- probabilistic networks
- maximum likelihood
- particle filter
- sensitivity analysis
- hidden variables
- markov chain monte carlo
- probabilistic graphical models
- expectation maximization
- graph structure
- bregman divergences
- posterior distribution
- latent variable models
- factor graphs
- mixture model
- em algorithm