Rademacher Averages


So far we have seen how to obtain generalization error bounds for learning algorithms that pick a function from a function class of limited capacity or complexity, where the complexity of the class is measured using the growth function or VC-dimension in the binary case, and using covering numbers or the fatshattering dimension in the real-valued case… (More)


Figures and Tables

Sorry, we couldn't extract any figures or tables for this paper.

Slides referencing similar topics