Improved N-Division Output Coding for Multiclass Learning Problems.
Jaepil KoEunju KimHyeran ByunPublished in: ICPR (3) (2004)
Keyphrases
- learning problems
- multi class
- binary classification
- multiclass learning
- error correcting output codes
- learning tasks
- multiclass classification
- machine learning algorithms
- multi task
- kernel methods
- semi supervised learning
- supervised learning
- learning algorithm
- multiclass problems
- multi class classification
- binary classifiers
- cost sensitive
- machine learning
- multi task learning
- support vector machine
- pairwise
- reinforcement learning
- binary classification problems
- feature selection
- sample complexity
- perceptron algorithm
- multiple kernel learning
- pac learning
- feature vectors
- data sets
- reproducing kernel hilbert space
- upper bound
- training set
- training data
- structured sparsity