An infinite-horizon robust filter for uncertain hidden Markov models with conditional relative entropy constraints.
Jason J. FordValery A. UgrinovskiiJohn LaiPublished in: AuCC (2011)
Keyphrases
- hidden markov models
- infinite horizon
- relative entropy
- optimal control
- long run
- optimal policy
- information theoretic
- information theory
- markov decision processes
- covariance matrix
- log likelihood
- conditional random fields
- dynamic programming
- state space
- discriminative training
- noise reduction
- mutual information
- machine learning
- maximum entropy
- random variables
- lead time
- generative model
- markov chain
- partially observable