Variational Inference on Infinite Mixtures of Inverse Gaussian, Multinomial Probit and Exponential Regression.
Sk. Minhazul IslamArunava BanerjeePublished in: ICMLA (2014)
Keyphrases
- variational inference
- mixture model
- exponential family
- maximum likelihood
- probabilistic model
- gaussian mixture model
- em algorithm
- gaussian process
- expectation maximization
- regression model
- gaussian processes
- model selection
- bayesian inference
- gaussian distribution
- latent variables
- hyperparameters
- generative model
- posterior distribution
- density estimation
- variational methods
- topic models
- latent dirichlet allocation
- graphical models
- parameter estimation
- log likelihood
- probabilistic graphical models
- language model
- dirichlet process
- approximate inference
- bayesian framework
- hidden variables
- closed form
- conditional random fields
- unsupervised learning
- maximum a posteriori
- markov networks
- text classification
- semi supervised
- bayesian networks
- posterior probability
- machine learning
- missing values
- statistical models
- support vector