Task decomposition and modular single-hidden-layer perceptron classifiers for multi-class learning problems.
Gao DaqiLi ChunxiaYang YunfanPublished in: Pattern Recognit. (2007)
Keyphrases
- learning problems
- multi class
- error correcting output codes
- learning algorithm
- machine learning algorithms
- binary classifiers
- perceptron algorithm
- binary classification
- learning tasks
- kernel methods
- machine learning
- supervised learning
- support vector
- support vector machine
- multi class classification
- training data
- multiclass problems
- generalization bounds
- feature selection
- multi task learning
- semi supervised learning
- multiple kernel learning
- multi task
- multi class problems
- reinforcement learning
- boosting algorithms
- cost sensitive
- neural network
- training examples
- training samples
- svm classifier
- pairwise
- decision trees
- classification algorithm
- active learning
- kernel function
- naive bayes
- base classifiers
- prior knowledge
- training set
- feature extraction
- data sets