Quasi-continuous maximum entropy distribution approximation with kernel density.
Thomas MazzoniElmar ReucherPublished in: Int. J. Inf. Decis. Sci. (2011)
Keyphrases
- maximum entropy
- kernel density
- mean shift
- maximum entropy principle
- joint distribution
- hidden markov models
- markov models
- class conditional
- random fields
- transformation based learning
- iterative scaling
- credal networks
- conditional random fields
- minimum cross entropy
- matrix approximation
- maximum entropy model
- feature space
- principle of maximum entropy
- image processing
- approximation error
- closed form
- maximum likelihood
- similarity measure