Exponentiated Gradient Algorithms for Conditional Random Fields and Max-Margin Markov Networks.
Michael CollinsAmir GlobersonTerry KooXavier CarrerasPeter L. BartlettPublished in: J. Mach. Learn. Res. (2008)
Keyphrases
- max margin
- exponentiated gradient
- markov networks
- conditional random fields
- structured prediction
- graphical models
- maximum margin
- belief propagation
- probabilistic model
- maximum likelihood
- markov random field
- information extraction
- generative model
- first order logic
- multi class
- bayesian inference
- hidden markov models
- bayesian networks
- latent variables
- conditional probabilities
- pairwise
- higher order
- maximum entropy
- probabilistic inference
- structure learning
- parameter learning
- belief networks
- document classification
- learning algorithm
- text classification
- computer vision
- log linear models
- keywords
- efficient learning
- hidden variables
- segmentation method
- inductive logic programming
- posterior probability