A bilevel learning approach for optimal observation placement in variational data assimilation

DOI10.1088/1361-6420/AB4BFAMaRDI QIDQ5000561zbMATH OpenOpenAlexFDO

Authors

Publication date 14 July 2021

Published in Inverse Problems (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1811.11505

zbMATH Keywords

bilevel optimizationsingular optimal control with measuressupervised learningvariational data assimilation

Mathematics Subject Classification ID

PDEs in connection with control and optimization (35Q93) Existence theories for optimal control problems involving partial differential equations (49J20) Optimality conditions for problems involving partial differential equations (49K20) Newton-type methods (49M15) PDE constrained optimization (numerical aspects) (49M41) Numerical solution to inverse problems in abstract spaces (65J22) Learning and adaptive systems in artificial intelligence (68T05) Large-scale problems in mathematical programming (90C06) Methods of quasi-Newton type (90C53)

Abstract: In this paper we propose a bilevel optimization approach for the placement of space and time observations in variational data assimilation problems. Within the framework of supervised learning, we consider a bilevel problem where the lower-level task is the variational reconstruction of the initial condition of a semilinear system, and the upper-level problem solves the optimal placement with help of a sparsity inducing function. Due to the pointwise nature of the observations, an optimality system with regular Borel measures is obtained as necessary optimality condition for the lower-level problem. The latter is then considered as constraint for the upper-level instance, yielding an optimization problem constrained by a multi-state system with measures. We demonstrate existence of Lagrange multipliers and derive a necessary optimality system characterizing the optimal solution of the bilevel problem. The numerical solution is carried out also on two levels. The lower-level problem is solved using a standard BFGS method, while the upper-level one is solved by means of a projected BFGS algorithm based on the estimation of $epsilon$-active sets. Finally some numerical experiments are presented to illustrate the main features of our approach.

Recommendations

• Bilevel approaches for learning of variational imaging models
• On optimal solution error covariances in variational data assimilation
• On optimal solution error covariances in variational data assimilation problems
• Sensitivity with respect to observations in variational data assimilation
• Optimal sensor placement for variational data assimilation of unsteady flows past a rotationally oscillating cylinder

Cites work

• A bilevel approach for parameter learning in inverse problems
• A Fast and Scalable Method for A-Optimal Design of Experiments for Infinite-dimensional Bayesian Nonlinear Inverse Problems
• A-optimal design of experiments for infinite-dimensional Bayesian linear inverse problems with regularized _0-sparsification
• Analysis in Banach Spaces
• Existence of solutions of a thermoviscoplastic model and associated optimal control problems
• scientific article; zbMATH DE number 3813420 (Why is no real title available?)
• scientific article; zbMATH DE number 3602126 (Why is no real title available?)
• scientific article; zbMATH DE number 3619756 (Why is no real title available?)
• scientific article; zbMATH DE number 1274356 (Why is no real title available?)
• scientific article; zbMATH DE number 1064079 (Why is no real title available?)
• scientific article; zbMATH DE number 3277871 (Why is no real title available?)
• scientific article; zbMATH DE number 3322278 (Why is no real title available?)
• scientific article; zbMATH DE number 3366787 (Why is no real title available?)
• scientific article; zbMATH DE number 3302617 (Why is no real title available?)
• Hölder estimates for parabolic operators on domains with rough boundary
• Image denoising: learning the noise model via nonsmooth PDE-constrained optimization
• Is bilevel programming a special case of a mathematical program with complementarity constraints?
• Nonautonomous parabolic equations involving measures
• Numerical methods for the design of large-scale nonlinear discrete ill-posed inverse problems
• Numerical PDE-constrained optimization
• Optimal control of nonsmooth, semilinear parabolic equations
• Optimal Control of Semilinear Multistate Systems with State Constraints
• Optimal Design of Experiments
• Optimal experimental design for inverse problems with state constraints
• Optimal sensor placement: a robust approach
• Optimal shape and location of sensors for parabolic equations with random initial data
• Parabolic control problems in measure spaces with sparse solutions
• Sequentially optimal sensor placement in thermoelastic models for real time applications
• State estimation for infinite-dimensional systems
• The infinite-dimensional optimal filtering problem with mobile and stationary sensor networks

Cited in

(6)

• Consistency analysis of bilevel data-driven learning in inverse problems
• An optimal control approach for determining the source term in fractional diffusion equation by different cost functionals
• Optimal experimental design for infinite-dimensional Bayesian inverse problems governed by PDEs: a review
• An optimal fluid optical flow registration for super-resolution with Lamé parameters learning
• Optimal sensor placement for ensemble-based data assimilation using gradient-weighted class activation mapping
• Deep learning for PDE-based inverse problems. Abstracts from the workshop held October 27 -- November 1, 2024

Describes a project that uses

Uses Software

• KELLEY

This page was built for publication: A bilevel learning approach for optimal observation placement in variational data assimilation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5000561)