Gaussian kernel c-means hard clustering algorithms with automated computation of the width hyper-parameters.
Francisco de A. T. de CarvalhoEduardo C. SimõesLucas V. C. SantanaMarcelo Rodrigo Portela FerreiraPublished in: Pattern Recognit. (2018)
Keyphrases
- gaussian kernel
- hyperparameters
- clustering algorithm
- model selection
- scale space
- support vector
- cross validation
- bayesian framework
- kernel methods
- random sampling
- bayesian inference
- closed form
- noise level
- support vector machine
- prior information
- em algorithm
- optimal parameters
- sample size
- gaussian process
- decision trees
- incremental learning
- data sets
- parameter space
- maximum a posteriori
- generative model
- probabilistic model
- prior knowledge
- missing values
- maximum likelihood
- active learning