Sample Size Issues in the Choice between the Best Classifier and Fusion by Trainable Combiners.
Sarunas RaudysGiorgio FumeraAistis RaudysIgnazio PillaiPublished in: IDEAL (2014)
Keyphrases
- sample size
- small sample
- individual classifiers
- upper bound
- model selection
- random sampling
- covariance matrix
- training data
- small sample size
- classifier combination
- progressive sampling
- feature selection
- experimental design
- pac learning
- number of training samples
- statistical power
- vc dimension
- lower bound
- support vector
- confidence intervals
- small samples
- variance reduction
- class labels
- feature space
- worst case
- generalization error
- classification algorithm
- training samples
- combining multiple