Learning Additive Exponential Family Graphical Models via \ell_{2, 1}-norm Regularized M-Estimation.
Xiao-Tong YuanPing LiTong ZhangQingshan LiuGuangcan LiuPublished in: NIPS (2016)
Keyphrases
- graphical models
- exponential family
- probabilistic model
- density estimation
- approximate inference
- markov networks
- maximum likelihood
- log likelihood
- random variables
- learning algorithm
- belief propagation
- probabilistic inference
- bregman divergences
- missing values
- belief networks
- closed form
- hidden variables
- conditional random fields
- conditional independence
- prior knowledge
- pairwise
- bayesian networks
- machine learning
- ell norm
- learning problems
- statistical models
- mixture model
- supervised learning
- least squares
- high dimensional
- feature space