Risk bounds for statistical learning
Pascal Massart, Élodie Nédélec · The Annals of Statistics · 2006
We propose a general theorem providing upper bounds for the risk of an empirical risk minimizer (ERM).We essentially focus on the binary classification framework. We extend Tsybakov’s analysis of the risk of an ERM under margin type conditions by using concentration inequalities for conveniently weighted empirical processes. This allows us to deal with ways of measuring the “size” of a class of classifiers other than entropy with bracketing as in Tsybakov’s work. In particular, we derive new risk bounds for the ERM when the classification rules belong to some VC-class under margin conditions and discuss the optimality of these bounds in a minimax sense.
Cited by 253 works. doi:10.1214/009053606000000786 Open-access PDF
Authors
References
- Estimation of Dependences Based on Empirical Data
- Algorithms in Combinatorial Geometry.
- Algorithms in Combinatorial Geometry
- On the method of bounded differences
- Probability in Banach Spaces: Isoperimetry and Processes
- New concentration inequalities in product spaces
- Minimax Theory of Image Reconstruction
- Optimal aggregation of classifiers in statistical learning
- Risk bounds for model selection via penalization
- Uniform Central Limit Theorems