Parameter-Free Online Learning via Model Selection.
Dylan J. FosterSatyen KaleMehryar MohriKarthik SridharanPublished in: NIPS (2017)
Keyphrases
- parameter free
- model selection
- online learning
- cross validation
- categorical data
- hyperparameters
- parameter estimation
- bayesian learning
- regression model
- outlier detection
- e learning
- selection criterion
- active learning
- marginal likelihood
- statistical inference
- unsupervised learning
- sample size
- variable selection
- mixture model
- machine learning
- information criterion
- error estimation
- automatic model selection
- fully automatic
- model selection criteria
- gaussian process
- bayesian model selection
- feature selection
- bayesian information criterion
- generalization error