Improving the sample complexity using global data

IEEE Transactions on Information Theory

Volumn 48, Issue 7, 2002, Pages 1977-1991

  Mendelson, Shahar ^a  

^a AUSTRALIAN NATIONAL UNIVERSITY   (Australia)

Author keywords

Fat shattering dimension; Glivenko Cantelli classes; Kernel machines; Learning sample complexity; Uniform convexity

Indexed keywords

APPROXIMATION THEORY; CONVERGENCE OF NUMERICAL METHODS; ESTIMATION; POLYNOMIALS; PROBABILITY; REGRESSION ANALYSIS; SET THEORY;

LEARNING PROBLEMS;

INFORMATION THEORY;

EID: 0036648803     PISSN: 00189448     EISSN: None     Source Type: Journal    
DOI: 10.1109/TIT.2002.1013137     Document Type: Article

References (21)

1. Metric entropy of convex hulls in Hilbert spaces
2. Metric entropy of convex hulls in Banach spaces
3. On the mathematical foundations of learning
4. Universal Donsker classes and metric entropy
5. A note on the scale-sensitive dimension of linear bounded functionals in Banach spaces
6. P
7. The importance of convexity in learning with squared loss
8. About the constants in Talagrand's concentration inequalities for empirical processes
9. Learnability in Hilhert spaces with reproducing kernels
10. Rademacher averages and phase transitions in Glivenko-Cantelli classes
11. On the size of convex hulls of small sets
12. Integral transforms, reproducing kernels and their applications
13. Sharper bounds for Gaussian and empirical processes
14. Generalization performance of regularization networks and support vectors machines via entropy numbers of compact operators