Open Access
August 2018 On optimality of empirical risk minimization in linear aggregation
Adrien Saumard
Bernoulli 24(3): 2176-2203 (August 2018). DOI: 10.3150/17-BEJ925


In the first part of this paper, we show that the small-ball condition, recently introduced by (J. ACM 62 (2015) Art. 21, 25), may behave poorly for important classes of localized functions such as wavelets, piecewise polynomials or for trigonometric polynomials, in particular leading to suboptimal estimates of the rate of convergence of ERM for the linear aggregation problem. In a second part, we recover optimal rates of convergence for the excess risk of ERM when the dictionary is made of trigonometric functions. Considering the bounded case, we derive the concentration of the excess risk around a single point, which is an information far more precise than the rate of convergence. In the general setting of a $L_{2}$ noise, we finally refine the small ball argument by rightly selecting the directions we are looking at, in such a way that we obtain optimal rates of aggregation for the Fourier dictionary.


Download Citation

Adrien Saumard. "On optimality of empirical risk minimization in linear aggregation." Bernoulli 24 (3) 2176 - 2203, August 2018.


Received: 1 October 2016; Published: August 2018
First available in Project Euclid: 2 February 2018

zbMATH: 06839264
MathSciNet: MR3757527
Digital Object Identifier: 10.3150/17-BEJ925

Keywords: empirical risk minimization , excess risk’s concentration , linear aggregation , Optimal rates , small-ball property

Rights: Copyright © 2018 Bernoulli Society for Mathematical Statistics and Probability

Vol.24 • No. 3 • August 2018
Back to Top