论文信息 - Localized Rademacher Complexities - 字舞流文

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.

Peter L. Bartlett | Shahar Mendelson | Olivier Bousquet | P. Bartlett | O. Bousquet | S. Mendelson

[1] Peter L. Bartlett,et al. Efficient agnostic learning of neural networks with bounded fan-in , 1996, IEEE Trans. Inf. Theory.

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

[3] V. Koltchinskii,et al. Rademacher Processes and Bounding the Risk of Function Learning , 2004, math/0405338.

[4] P. Massart. Some applications of concentration inequalities to statistics , 2000 .

[5] Peter L. Bartlett,et al. Rademacher and Gaussian Complexities: Risk Bounds and Structural Results , 2003, J. Mach. Learn. Res..

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

[7] Shahar Mendelson,et al. Agnostic Learning Nonconvex Function Classes , 2002, COLT.

[8] Shahar Mendelson,et al. Improving the sample complexity using global data , 2002, IEEE Trans. Inf. Theory.

[9] P. MassartLedoux,et al. Concentration Inequalities Using the Entropy Method , 2002 .

[10] E. Rio. Une inégalité de Bennett pour les maxima de processus empiriques , 2002 .

[11] O. Bousquet. A Bennett concentration inequality and its application to suprema of empirical processes , 2002 .

[12] Peter L. Bartlett,et al. Model Selection and Error Estimation , 2000, Machine Learning.