Heterogeneous features and model selection for event-based media classification.
Xueliang LiuBenoit HuetPublished in: ICMR (2013)
Keyphrases
- model selection
- cross validation
- hyperparameters
- machine learning
- statistical learning
- sample size
- regression model
- feature selection
- bayesian methods
- parameter estimation
- variable selection
- error estimation
- meta learning
- mixture model
- model selection criteria
- classification accuracy
- leave one out cross validation
- bayesian learning
- selection criterion
- statistical inference
- unsupervised learning
- generalization error
- gaussian process
- feature vectors
- marginal likelihood
- generalization bounds
- bayesian model selection
- decision trees
- support vector
- training set
- data sets
- maximum likelihood