Choice and Value Flexibility Jointly Contribute to the Capacity of a Subsampled Quadratic Classifier.
Panayiota PoiraziBartlett W. MelPublished in: Neural Comput. (2000)
Keyphrases
- spatial resolution
- feature selection
- classification scheme
- classifier systems
- training data
- feature space
- support vector machine
- objective function
- pairwise
- feature set
- learning algorithm
- classification rate
- spatial information
- nearest neighbor
- training set
- computational complexity
- support vector
- decision trees
- data sets
- image classification
- classification method
- linear classifiers
- multiple kernel learning
- binary classifiers
- spatial frequency
- nearest neighbor classifier