Kernel-based MinMax clustering methods with kernelization of the metric and auto-tuning hyper-parameters.
Junyan LiuYongan GuoDapeng LiZefeng WangYouyun XuPublished in: Neurocomputing (2019)
Keyphrases
- hyperparameters
- support vector
- model selection
- closed form
- cross validation
- parameter settings
- bayesian inference
- bayesian framework
- random sampling
- prior information
- gaussian process
- variational bayes
- em algorithm
- noise level
- incremental learning
- maximum likelihood
- maximum a posteriori
- support vector machine
- posterior distribution
- sample size
- incomplete data
- missing values
- bayesian methods
- active learning
- decision trees
- e learning
- expectation maximization
- probabilistic model
- multiscale
- data sets