Remarks on Talagrand’s deviation inequality for Rademacher functions
نویسنده
چکیده
where σ is the Lipschitz constant of the extension of f and P is the natural probability on {0, 1}. Here we extend this inequality to more general product probability spaces; in particular, we prove the same inequality for {0, 1} with the product measure ((1− η)δ0 + ηδ1) . We believe this should be useful in proofs involving random selections. As an illustration of possible applications we give a simple proof (though not with the right dependence on ε) of the Bourgain, Lindenstrauss, Milman result [BLM] that for 1 ≤ r < s ≤ 2 and ε > 0, every n-dimensional subspace of Ls (1+ ε)embeds into lr with N = c(r, s, ε)n. The main results. For i = 1, . . . , n let (Xi, ‖ · ‖i), be normed spaces, let Ωi be a finite subset of Xi with diameter at most one and let Pi be a probability measure on Ωi. Define
منابع مشابه
Learning Theory and Algorithms for Revenue Optimization in Second-Price Auctions with Reserve A. Proofs for learning guarantees
A.2. Contraction lemma The following is a version of Talagrand’s contraction lemma (Ledoux & Talagrand, 2011). Since our definition of Rademacher complexity does not use absolute values, we give an explicit proof below. Lemma 8. Let H be a hypothesis set of functions mapping X to R and Ψ1, . . . ,Ψm, μ-Lipschitz functions for some μ > 0. Then, for any sample S of m points x1, . . . , xm ∈ X , t...
متن کاملMoment inequalities for functions of independent random variables
A general method for obtaining moment inequalities for functions of independent random variables is presented. It is a generalization of the entropy method which has been used to derive concentration inequalities for such functions [7], and is based on a generalized tensorization inequality due to Lata la and Oleszkiewicz [25]. The new inequalities prove to be a versatile tool in a wide range o...
متن کاملMoment Inequalities for Functions of Independent Random Variables by Stéphane Boucheron,1 Olivier Bousquet,
A general method for obtaining moment inequalities for functions of independent random variables is presented. It is a generalization of the entropy method which has been used to derive concentration inequalities for such functions [Boucheron, Lugosi and Massart Ann. Probab. 31 (2003) 1583–1614], and is based on a generalized tensorization inequality due to Latała and Oleszkiewicz [Lecture Note...
متن کاملSymmetrization Approach to Concentration Inequalities for Empirical Processes
We introduce a symmetrization technique that allows us to translate a problem of controlling the deviation of some functionals on a product space from their mean into a problem of controlling the deviation between two independent copies of the functional. As an application we give a new easy proof of Talagrand’s concentration inequality for empirical processes, where besides symmetrization we u...
متن کاملRademacher Complexity
Rademacher complexity is a measure of the richness of a class of real-valued functions. In this sense, it is similar to the VC dimension. In fact, we will establish a uniform deviation bound in terms of Rademacher complexity, and then use this result to prove the VC inequality. Unlike VC dimension, however, Rademacher complexity is not restricted to binary functions, and will also prove useful ...
متن کامل