Univ. Heidelberg
Statistics Group   Institute for Mathematics   Faculty of Mathematics and Computer Science   University Heidelberg
Ruprecht-Karls-Universität Heidelberg Institute for Mathematics Statistics of inverse problems Research Group
german english french



Publications
Cooperations
Research projects
Events
Teaching
Completed theses
People
Contact


Last edited on
Oct 17, 2024 by JJ
.
Thesis:
Bachelor in Mathematics

Author:
Robin Viellieber

Title:
Optimal rates of aggregation in classification

Supervisor:
Jan JOHANNES

Abstract:
Using an aggregation procedure, this thesis aims to establish optimal rates of convergence for four aggregates in the context of classification. The here stated results rely on a low noise assumption and are provided with respect to the Bayes rule. However, we will construct the theory using the convex hinge loss and then derive similar results for the Bayes rule. Moreover, we introduce an approach to optimality based on Tsybakov (2003).

References:
G. Lecué. Optimal rates of aggregation in classification under low noise assumption. Bernoulli, 13(4):1000-1022, 2007.
A. B. Tsybakov. Optimal rates of aggregation. In: Schölkopf, Bernhard (ed.) et al., Learning theory and kernel machines. 16th annual conference on learning theory and 7th Kernel workshop proceedings. Springer Lecture Notes Computer Science, 2777, 303-313, 2003.