Observability as a constraint qualification for a nonsmooth bilevel PDE-constrained problem
Abstract
Motivated by experiment design for parameter identification in image reconstruction, we analyze a bilevel control problem involving sparse inversion at the lower level. Qualified optimality conditions in explicit multiplier form are shown to hold under either an observability condition for a coupled system or the Lebesgue nullity of a contact set induced by the lower-level adjoint variable. We prove that these conditions are independent of one another and that they cannot, in general, both be omitted. The approach is based on the characterization of local first-order geometric objects (tangent and normal cones) associated with a set of lifted constraints. Based on the resulting optimality conditions, we formulate a reduced gradient method and illustrate it through numerical experiments.
Related
- The Unified Separation Condition: A General Constraint Qualification for Smooth and Nonsmooth Optimization
- Bi-ZOL: Bilevel Zeroth-Order Learning with Nonsmooth Responses
- Nonmaximal sums of maximally monotone operators under Rockafellar's constraint qualification
- Dynamic Regret and Tightened Constraint Violation for Communication-Efficient Distributed Online Bilevel Optimization
- Discrete-time feedback linearization control for nonsmooth constrained optimization
BibTeX
@misc{clason2026observability,
title = {{Observability as a constraint qualification for a nonsmooth bilevel PDE-constrained problem}},
author = {Christian Clason and Alberto Domínguez Corella and Jyrki Jauhiainen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.21377},
eprint = {2610.21377},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.21377},
}
RIS
TY - UNPB
TI - Observability as a constraint qualification for a nonsmooth bilevel PDE-constrained problem
AU - Christian Clason
AU - Alberto Domínguez Corella
AU - Jyrki Jauhiainen
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.21377
UR - https://arxiv.org/abs/2610.21377
AB - Motivated by experiment design for parameter identification in image reconstruction, we analyze a bilevel control problem involving sparse inversion at the lower level. Qualified optimality conditions in explicit multiplier form are shown to hold under either an observability condition for a coupled system or the Lebesgue nullity of a contact set induced by the lower-level adjoint variable. We prove that these conditions are independent of one another and that they cannot, in general, both be omitted. The approach is based on the characterization of local first-order geometric objects (tangent and normal cones) associated with a set of lifted constraints. Based on the resulting optimality conditions, we formulate a reduced gradient method and illustrate it through numerical experiments.
ER -