Local Rademacher Complexities

نویسندگان

  • Peter L. Bartlett
  • Olivier Bousquet
چکیده

We propose new bounds on the error of learning algorithms in terms of a data-dependent notion of complexity. The estimates we establish give optimal rates and are based on a local and empirical version of Rademacher averages, in the sense that the Rademacher averages are computed from the data, on a subset of functions with small empirical error. We present some applications to classification, prediction with a convex function class and in particular kernel classes.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Local Rademacher complexity bounds based on covering numbers

This paper provides a general result on controlling local Rademacher complexities, which captures in an elegant form to relate the complexities with constraint on the expected norm to the corresponding ones with constraint on the empirical norm. This result is convenient to apply in real applications and could yield refined local Rademacher complexity bounds for function classes satisfying gene...

متن کامل

Localized Rademacher Complexities

We investigate the behaviour of global and local Rademacher averages. We present new error bounds which are based on the local averages and indicate how data-dependent local averages can be estimated without a priori knowledge of the class at hand.

متن کامل

Rejoinder: 2004 Ims Medallion Lecture: Local Rademacher Complexities and Oracle Inequalities in Risk Minimization

of the true risk function F ∋ f 7→ Pf. The first quantity of interest is the L2-diameter of this set, D(F ; δ), and the second one is the function φn(F ; δ) that is equal to the expected supremum of empirical process indexed by the differences f − g, f, g ∈F(δ). These two functions are then combined in the expression Ūn(δ; t) that has its roots in Talagrand’s concentration inequalities for empi...

متن کامل

Localization and Adaptation in Online Learning

We introduce a formalism of localization for online learning problems, which, similarly to statistical learning theory, can be used to obtain fast rates. In particular, we introduce local sequential Rademacher complexities and other local measures. Based on the idea of relaxations for deriving algorithms, we provide a template method that takes advantage of localization. Furthermore, we build a...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2002