New Liftable Classes for First-Order Probabilistic Inference.
Seyed Mehran KazemiAngelika KimmigGuy Van den BroeckDavid PoolePublished in: NIPS (2016)
Keyphrases
- probabilistic inference
- graphical models
- bayesian networks
- conditional probabilities
- efficient inference
- influence diagrams
- weighted model counting
- message passing
- approximate inference
- context specific independence
- belief networks
- higher order
- probabilistic reasoning
- bayesian belief networks
- variable elimination
- first order logic
- exact inference
- class labels
- semi supervised learning
- bayesian inference
- classification accuracy
- probabilistic model
- prior knowledge