Stopping criterion for active learning based on deterministic generalization bounds.
Hideaki IshibashiHideitsu HinoPublished in: CoRR (2020)
Keyphrases
- generalization bounds
- stopping criterion
- active learning
- statistical learning theory
- learning theory
- generalization ability
- data dependent
- vc dimension
- convergence rate
- learning problems
- learning algorithm
- model selection
- sample complexity
- linear classifiers
- cost function
- ranking algorithm
- supervised learning
- semi supervised
- ranking functions
- machine learning
- generalization error
- theoretical framework
- training examples
- random sampling
- region merging
- learning machines
- learning process
- semi supervised learning
- kernel machines
- support vector machine svm
- training set
- statistical learning
- unlabeled data
- relevance feedback
- supervised classification
- support vector machine
- feature selection
- upper bound