A New Geometric Approach to the Complexity of Model Selection.
Ziang LvSiwei LuoYunhui LiuYu ZhengPublished in: IEEE ICCI (2006)
Keyphrases
- model selection
- cross validation
- bayesian learning
- hyperparameters
- regression model
- parameter estimation
- error estimation
- selection criterion
- meta learning
- statistical learning
- model selection criteria
- sample size
- machine learning
- variable selection
- mixture model
- motion segmentation
- generalization error
- feature selection
- statistical inference
- marginal likelihood
- worst case
- bayesian methods
- parameter determination
- generalization bounds
- bayesian information criterion
- information criterion
- automatic model selection
- data mining
- active learning
- semi supervised
- closed form