On Concentration of Discrete Distributions with Applications to Supervised Learning of Classifiers.
Magnus EkdahlTimo KoskiPublished in: MLDM (2007)
Keyphrases
- supervised learning
- training data
- supervised classification
- training set
- class labels
- training examples
- training samples
- labeled training data
- co training
- unsupervised learning
- efficient algorithms to compute
- accurate classification
- probability distribution
- support vector
- learning algorithm
- active learning
- supervised training
- random variables
- supervised learning algorithms
- instance selection
- class conditional
- classification algorithm
- machine learning
- semi supervised learning
- svm classifier
- learning problems
- classification models
- feature selection
- reinforcement learning
- unlabeled data
- multiple instance learning
- test set
- data sets
- class conditional densities
- decision trees
- labeled data
- continuous domains
- classification rate
- power law
- learning machines
- linear classifiers
- semi supervised
- machine learning algorithms
- generalization error
- learning tasks
- machine learning methods