Rademacher averages and phase transitions in Glivenko-Cantelli classes

IEEE Transactions on Information Theory

Volumn 48, Issue 1, 2002, Pages 251-263

  Mendelson, Shahar ^a  

^a AUSTRALIAN NATIONAL UNIVERSITY   (Australia)

Author keywords

Fat shattering dimension; Rademacher averages; Uniform Glivenko Cantelli (GC) classes

Indexed keywords

APPROXIMATION THEORY; COMPUTATIONAL COMPLEXITY; PARAMETER ESTIMATION; PROBABILITY; THEOREM PROVING;

FAT SHATTERING DIMENSIONS; GLIVENKO-CANTELLI CLASSES; RADEMACHER AVERAGES;

INFORMATION THEORY;

EID: 0036160893     PISSN: 00189448     EISSN: None     Source Type: Journal    
DOI: 10.1109/18.971753     Document Type: Article

References (23)

1. Scale sensitive dimensions, uniform convergence and learnability
2. Covering numbers for real valued function classes
3. More theorems about scale sensitive dimensions and learning
4. The sizes of compact subsets of Hilbert space and continuity of Gaussian processes
5. Uniform central limit theorems
6. Uniform and universal Glivenko-Cantelli classes
7. n
8. Empirical processes and applications: An overview
9. Some limit theorems for empirical processes
10. Agnostic learning and single hidden layer neural networks
11. The importance of convexity in learning with squared loss
12. The local theory of normed spaces and its application to convexity
13. Geometric methods in the analysis of Glivenko-Cantelli classes
14. Improving the sample complexity using global data
15. On the geometry of Glivenko-Cantelli classes
16. Gaussian processes and measures of solid angles in Hilbert space
17. Type, infratype and the Elton-Pajor theorem
18. Sharper bounds for Gaussian and empirical processes
19. Banach-Mazur distance and finite-dimensional operator ideals
20. Necessary and sufficient conditions for uniform convergence of means to mathematical expectations