Hyper-sparse optimal aggregation

Abstract

In this paper, we consider the problem of "hyper-sparse aggregation". Namely, given a dictionary $F=\{f_1,...,f_M\}$ of functions, we look for an optimal aggregation algorithm that writes $\tilde{f}={\sum }_{j=1}^M{\theta }_j{f}_j$ with as many zero coefficients ${\theta }_j$ as possible. This problem is of particular interest when $F$ contains many irrelevant functions that should not appear in $\tilde{f}$. We provide an exact oracle inequality for $\tilde{f}$, where only two coefficients are non-zero, that entails $\tilde{f}$ to be an optimal aggregation algorithm. Since selectors are suboptimal aggregation procedures, this proves that 2 is the minimal number of elements of $F$ required for the construction of an optimal aggregation procedures in every situations. A simulated example of this algorithm is proposed on a dictionary obtained using LARS, for the problem of selection of the regularization parameter of the LASSO. We also give an example of use of aggregation to achieve minimax adaptation over anisotropic Besov spaces, which was not previously known in minimax theory (in regression on a random design).

33 pages

References in corpus (1)

• Fast learning rates in statistical inference through aggregation

Cited by in corpus (5)

• Deviation optimal learning using greedy Q-aggregation
• Deep learning is adaptive to intrinsic dimensionality of model smoothness in anisotropic Besov space
• A tutorial on estimator averaging in spatial point process models
• Bayesian Model Averaging with Exponentiated Least Square Loss
• Estimation of conditional cumulative distribution function from current status data