Message passing for Hybrid Bayesian Networks using Gaussian mixture reduction.
Cheol Young ParkKathryn Blackmond LaskeyPaulo C. G. CostaShou MatsumotoPublished in: ICDIM (2015)
Keyphrases
- message passing
- gaussian mixture
- bayesian networks
- belief propagation
- probabilistic inference
- junction tree
- expectation maximization
- em algorithm
- graphical models
- distributed systems
- approximate inference
- shared memory
- covariance matrix
- sum product algorithm
- gaussian mixture model
- closed form
- probability density function
- density estimation
- mixture model
- inference in graphical models
- markov random field
- probabilistic reasoning
- probabilistic model
- hidden variables
- sum product
- object recognition
- reinforcement learning
- conditional probabilities
- image patches
- multiscale
- high quality
- unsupervised learning