论文信息 - Rademacher Processes and Bounding the Risk of Function Learning

Rademacher Processes and Bounding the Risk of Function Learning

We construct data dependent upper bounds on the risk in function learning problems. The bounds are based on local norms of the Rademacher process indexed by the underlying function class, and they do not require prior knowledge about the distribution of training examples or any specific properties of the function class. Using Talagrand’s type concentration inequalities for empirical and Rademacher processes, we show that the bounds hold with high probability that decreases exponentially fast when the sample size grows. In typical situations that are frequently encountered in the theory of function learning, the bounds give nearly optimal rate of convergence of the risk to zero.

V. Koltchinskii | D. Panchenko

[1] E. Mammen,et al. Asymptotical minimax recovery of sets with smooth boundaries , 1995 .

[2] László Györfi,et al. A Probabilistic Theory of Pattern Recognition , 1996, Stochastic Modelling and Applied Probability.

[3] Jon A. Wellner,et al. Weak Convergence and Empirical Processes: With Applications to Statistics , 1996 .

[4] M. Talagrand. New concentration inequalities in product spaces , 1996 .

[5] M. Talagrand. Majorizing measures: the generic chaining , 1996 .

[6] M. Talagrand. A new look at independence , 1996 .

[7] P. Massart,et al. From Model Selection to Adaptive Estimation , 1997 .

[8] Mathukumalli Vidyasagar,et al. A Theory of Learning and Generalization , 1997 .

[9] Vladimir Vapnik,et al. Statistical learning theory , 1998 .

[10] P. Massart,et al. Risk bounds for model selection via penalization , 1999 .

[11] Statistical Learning Control of Uncertain Systems : It is Better Than It Seems , 1999 .

[12] Fernando Lozano,et al. Model selection using Rademacher Penalization , 2000 .

[13] P. Massart,et al. About the constants in Talagrand's concentration inequalities for empirical processes , 2000 .

[14] V. Koltchinski. Statistical Learning Control of Uncertain Systems: It is Better Than It Seems , 2000 .

[15] Vladimir Koltchinskii,et al. Rademacher penalties and structural risk minimization , 2001, IEEE Trans. Inf. Theory.

[16] A. W. van der Vaart,et al. Uniform Central Limit Theorems , 2001 .