Optimal design for linear models via gradient flow
Top Cited Papers
Open Access
- 1 January 2026
- journal article
- Published by American Institute of Mathematical Sciences (AIMS) in Communications on Pure and Applied Analysis
- Vol. 32, 282-311
Abstract
Optimal experimental design (OED) aims to choose the observations in an experiment to be as informative as possible, according to certain statistical criteria. In the linear case (when the observations depend linearly on the unknown parameters), it seeks the optimal weights over rows of the design matrix $ {\mathbf{A}} $ under certain criteria. Classical OED assumes a discrete design space and thus a design matrix with finite dimensions. In many practical situations, however, the design space is continuous-valued, so that the OED problem is one of optimizing over a continuous-valued design space. The objective becomes a functional over the probability measure, instead of a function of a finite dimensional vector. This change of perspective requires a new set of techniques to optimize over probability measures, and Wasserstein gradient flow becomes a natural candidate. Both the first-order criticality and the convexity properties of the OED objective are presented. Computationally, the Monte Carlo particle method is used to translate the gradient flow equation formulation into a numerical algorithm. This algorithm is applied to two elliptic inverse problems.
Keywords
All Related Versions
This publication has 38 references indexed in Scilit:
- Randomized numerical linear algebra: Foundations and algorithmsActa Numerica, 2020
- A sparse control approach to optimal sensor placement in PDE-constrained parameter estimation problemsNumerische Mathematik, 2019
- Nonlinear Inverse Problems2019
- Numerical Methods for Ordinary Differential Equations2016
- Equivalence of gradient flows and entropy solutions for singular nonlocal interaction equations in 1DESAIM: Control, Optimisation and Calculus of Variations, 2015
- Optimal Transport for Applied Mathematicians2015
- On the rate of convergence in Wasserstein distance of the empirical measureProbability Theory and Related Fields, 2014
- Convex Optimization2004
- Optimisation in space of measures and optimal designESAIM: Probability and Statistics, 2004
- Electrical impedance tomographyInverse Problems, 2002