VC-Dimension

From MahiragaWiki

Basic Definition

The theroy of uniform convergence of the Remp(w) to R(w) include bounds on the rate of convergence, which are based on Vapnik-Chervonenkis dimension (VC-Dimension). The VC-Dimension is a meausre of the capacity of the family of classification functions realized by the learning machine.

Simplify

Let the family of dichotomies implemented by learning machine <math>F = {F(x,w): wF, F:Rp->{0,1}}</math>