Abstract:
Abstract. In this article we introduce a new combinatorial parameter which generalizes the VC dimension and the fat-shattering dimension, and extends beyond the function-class setup. Using this parameter we establish entropy bounds for subsets of the n-dimensional unit cube, and in particular, we present new bounds on the empirical covering numbers and gaussian averages associated with classes of functions in terms of the fat-shattering dimension. 1
Citations
|
163
|
Scale-sensitive dimensions, uniform convergence, and learnability
– Alon, Ben-David, et al.
- 1997
|
|
122
|
Probability inequalities for sums of bounded random variables
– Hoeding
- 1963
|
|
118
|
The volume of convex bodies and Banach space geometry
– Pisier
- 1989
|
|
85
|
Sharper bounds for Gaussian and empirical processes
– Talagrand
- 1994
|
|
70
|
Sphere packing numbers for subsets of the Boolean n-cube with bounded Vapnik-Chervonenkis dimension
– Haussler
- 1995
|
|
59
|
Uniform Central Limit Theorems
– Dudley
- 1999
|
|
38
|
Improving the sample complexity using global data
– Mendelson
- 2002
|
|
24
|
Rademacher averages and phase transitions in Glivenko-Cantelli classes
– Mendelson
- 2002
|
|
15
|
Prediction, learning, uniform convergence, and scale-sensitive dimensions
– Bartlett, Long
- 1998
|
|
9
|
Chervonenkis: Necessary and sucient conditions for uniform convergence of means to mathematical expectations, Theory Prob
– Vapnik, A
- 1971
|
|
5
|
Geometric parameters of kernel machines
– Mendelson
- 2002
|
|
5
|
Sous espaces ` n 1 des espaces de Banach
– Pajor
- 1985
|
|
4
|
Gaussian charachterization of uniform Donsker classes of functions
– Giné, Zinn
- 1991
|
|
4
|
Asymptotic theory of dimensional normed spaces
– Milman, Schechtman
- 1986
|
|
