Designing Nonlinear Classifiers Through Minimizing VC Dimension Bound.
Jianhua XuPublished in: ISNN (1) (2005)
Keyphrases
- vc dimension
- generalization bounds
- upper bound
- learning machines
- lower bound
- vapnik chervonenkis
- risk bounds
- sample complexity
- sample size
- distribution free
- inductive inference
- concept classes
- worst case
- compression scheme
- statistical learning theory
- empirical risk minimization
- training data
- linear classifiers
- uniform convergence
- machine learning algorithms
- training examples
- pac learning
- support vector
- concept class
- training set
- feature selection
- decision trees
- multi class
- theoretical analysis
- data dependent
- training samples
- machine learning
- image compression
- supervised classification
- function classes
- generalization ability
- euclidean space
- loss function