Efficient Regularization Parameter Selection for Latent Variable Graphical Models via Bi-Level Optimization.
Joachim GiesenFrank NussbaumChristopher SchneiderPublished in: IJCAI (2019)
Keyphrases
- parameter selection
- graphical models
- bi level
- latent variables
- probabilistic model
- random variables
- adaptive regularization
- approximate inference
- probabilistic graphical models
- factor graphs
- belief propagation
- bayesian networks
- gray scale
- latent variable models
- model selection
- structure learning
- hidden variables
- markov networks
- ls svm
- image processing
- bayesian inference
- generative model
- statistical relational learning
- color images