Rademacher processes and bounding the risk of function learning

2004/05/17 by Vladimir Koltchinskii, Dmitry Panchenko · 2 citations
Mathematics · #math.PR #msc:60G35

paper · pdf

published as 2000 In: High Dimensional Probability II pp. 443 - 459 · 14 pages

arxiv created 2004/05/17 · arxiv updated 2009/12/01

Abstract

We construct data dependent bounds on the risk in function learning problems. The bounds are based on the local norms of the Rademacher process indexed by the underlying function class and they do not require prior knowledge about the distribution of the training examples or any specific properties of the function class.

Cited by

• Hardness, Structural Knowledge, and Opportunity: An Analytical Framework for Modular Performance Modeling
• Overfitting has a limitation: a model-independent generalization gap bound based on Rényi entropy

Related

• Risk Bounds and Rademacher Complexity in Batch Reinforcement Learning
• Rademacher learning rates for iterated random functions
• Uniform Risk Bounds for Learning with Dependent Data Sequences
• Local Rademacher Complexity-based Learning Guarantees for Multi-Task Learning
• Do we really need the Rademacher complexities?