Generalization performance of regularization networks and support vector machines via entropy numbers of compact operators

IEEE Transactions on Information Theory

Volumn 47, Issue 6, 2001, Pages 2516-2532

  Williamson, Robert C ^a,b   Smola, Alex J ^b   Schölkopf, Bernhard ^c  

^a IEEE   (Australia)

^b AUSTRALIAN NATIONAL UNIVERSITY   (Australia)

^c Barnhill Technologies   (United States)

Author keywords

entropy; Covering numbers; Kernel methods; Linear operators; Metric entropy; Statistical learning theory; Support vector (SV) machines

Indexed keywords

KERNEL METHODS; LINEAR OPERATORS; METRIC ENTROPY; REGULARIZATION NETWORK; STATISTICAL LEARNING THEORY; SUPPORT VECTOR MACHINES;

APPROXIMATION THEORY; EIGENVALUES AND EIGENFUNCTIONS; FUNCTIONS; MATHEMATICAL OPERATORS; PATTERN RECOGNITION; PROBLEM SOLVING; REGRESSION ANALYSIS; THEOREM PROVING;

LEARNING ALGORITHMS;

EID: 0035441827     PISSN: 00189448     EISSN: None     Source Type: Journal    
DOI: 10.1109/18.945262     Document Type: Article

References (62)

1. Theoretical foundations of the potential function method in pattern recognition learning
2. Characterization of ε-entropy in Gaussian processes
3. The asymptotic behavior of ε-entropy of a compact positive operator
4. Scale-sensitive dimensions, uniform convergence, and learnability
5. Probabilistic analysis of learning in artificial neural networks: The PAC model and its variants
6. Generalization performance of support vector machines and other pattern classifiers
7. Fat-shattering and the learnability of real-valued functions
8. A training algorithm for optimal margin classifiers
9. Entropy numbers of diagonal operators with an application to eigenvalue problems
10. Inequalities of Bernstein-Jackson-type and the degree of compactness of operators in Banach spaces
11. Support vector networks
12. Characterization of weak type by the entropy distribution of r-nuclear operators
13. Priors, stabilizers and basis functions: From regularization to radial, tensor and additive splines
14. Geometric and probabilistic estimates for entropy and approximation numbers of operators
15. Covering numbers for support vector machines
16. A note on a scale-sensitive dimension of linear bounded functionals in Banach spaces
17. ε-entropy and approximation of bandlimited functions
18. Some estimates of entropy numbers
19. Toward efficient agnostic learning
20. Operators of type p and metric entropy
21. Entropic order of operators in Banach spaces and the central limit theorem
22. ε-entropy and ε-capacity of sets in functional spaces
23. ε-entropy ε-rate, and interpolation spaces revisited with an application to linear communication channels
24. Learning pattern classification - A survey
25. The importance of convexity in learning with squared loss
26. Analysis of spherical symmetries in Euclidean spaces
27. Sur une propriété métrique de la dimension
28. The ε-entropy and ε-capacity of certain time-varying channels
29. The ε-entropy and ε-capacity of certain time-invariant channels
30. Nonlinear component analysis as a kernel eigenvalue problem
31. A mathematical theory of communication
32. A framework for structural risk minimization
33. Structural risk minimization over data-dependent hierarchies
34. Generalization performance of classifiers in terms of observed covering numbers
35. On a kernel-based method for pattern recognition, regression, approximation and operator inversion
36. Entropy numbers for convex combinations and mlps
37. The connection between regularization operators and support vector kernels
38. Regularized principal manifolds
39. The Glivenko-Cantelli problem, ten years later
40. Interpolationseigenschaften von Entropie- und Durchmesseridealen kompakter Operatoren
41. Necessary and sufficient conditions for the uniform convergence of means to their expectations
42. Asymptotic behavior of eigenvalues of certain integral operators
43. A maximum margin miscellany
44. Sample based generalization bounds
45. Entropy numbers of linear function classes