Simultaneous Model Selection and Optimization through Parameter-free Stochastic Learning.
Francesco OrabonaPublished in: NIPS (2014)
Keyphrases
- model selection
- parameter free
- statistical learning
- bayesian learning
- unsupervised learning
- cross validation
- meta learning
- learning machines
- sample size
- learning algorithm
- kernel machines
- hyperparameters
- parameter determination
- machine learning
- information criterion
- error estimation
- active learning
- reinforcement learning
- incremental learning
- learning tasks
- regression model
- selection criterion
- variable selection
- generalization error
- fully automatic
- bayesian methods
- mutual information
- supervised learning
- prior knowledge
- model selection criteria
- similarity measure