DeepAI AI Chat
Log In Sign Up

Refined bounds for randomized experimental design

by   Geovani Rizk, et al.

Experimental design is an approach for selecting samples among a given set so as to obtain the best estimator for a given criterion. In the context of linear regression, several optimal designs have been derived, each associated with a different criterion: mean square error, robustness, etc. Computing such designs is generally an NP-hard problem and one can instead rely on a convex relaxation that considers probability distributions over the samples. Although greedy strategies and rounding procedures have received a lot of attention, straightforward sampling from the optimal distribution has hardly been investigated. In this paper, we propose theoretical guarantees for randomized strategies on E and G-optimal design. To this end, we develop a new concentration inequality for the eigenvalues of random matrices using a refined version of the intrinsic dimension that enables us to quantify the performance of such randomized strategies. Finally, we evidence the validity of our analysis through experiments, with particular attention on the G-optimal design applied to the best arm identification problem for linear bandits.


page 1

page 2

page 3

page 4


Approximate Supermodularity Bounds for Experimental Design

This work provides performance guarantees for the greedy solution of exp...

A Randomized Exchange Algorithm for Computing Optimal Approximate Designs of Experiments

We propose a class of subspace ascent methods for computing optimal appr...

Computing optimal experimental designs with respect to a compound Bayes risk criterion

We consider the problem of computing optimal experimental design on a fi...

Ascent with Quadratic Assistance for the Construction of Exact Experimental Designs

In the area of experimental design, there is a large body of theoretical...

Model-assisted design of experiments in the presence of network correlated outcomes

We consider the problem of how to assign treatment in a randomized exper...

Estimating a scalar log-concave random variable, using a silence set based probabilistic sampling

We study the probabilistic sampling of a random variable, in which the v...

Experimental Design via Generalized Mean Objective Cost of Uncertainty

The mean objective cost of uncertainty (MOCU) quantifies the performance...