Finite Sample Behaviour of an Ergodically Fast Line-Search Algorithm.
Luc PronzatoHenry P. WynnAnatoly A. ZhigljavskyPublished in: Comput. Optim. Appl. (1999)
Keyphrases
- finite sample
- search algorithm
- sample size
- statistical learning theory
- uniform convergence
- nearest neighbor
- error bounds
- parzen window
- generalization bounds
- generalization error
- theoretical framework
- feature extraction
- cross validation
- reproducing kernel hilbert space
- statistical learning
- learning problems
- data sets
- model selection
- support vector machine
- similarity measure
- machine learning