A unifying approach to the shape and change-point hypotheses in the discrete univariate exponential family.
Chihiro HirotsuShoichi YamamotoHarukazu TsurutaPublished in: Comput. Stat. Data Anal. (2016)
Keyphrases
- exponential family
- change point
- maximum likelihood
- change point detection
- statistical models
- graphical models
- closed form
- density estimation
- log likelihood
- missing values
- mixture model
- variational methods
- order statistics
- hidden variables
- probability density function
- markov chain monte carlo
- feature extraction
- knowledge discovery
- evolutionary algorithm
- video sequences