POST-DOC POSITION (24 MONTHS) - POLYNOMIAL OPTIMIZATION FOR NETWORK
DYNAMICS
Location: LAAS-CNRS, Toulouse, France
Team: Polynomial Optimization (POP)
Project: ANR-DFG MONET (Moment-sum-of-squares Optimization for NETwork
dynamics)
Start Date: Early 2026
PROJECT DESCRIPTION:
Join the MONET project, a Franco-German collaboration (LAAS-CNRS & FAU
Erlangen-Nürnberg).
We are developing new methods to overcome the scalability barriers of
the moment-SOS hierarchy.
The project focuses on networked dynamical systems governed by nonlinear
hyperbolic PDEs,
which are critical for applications like gas pipelines, traffic flow,
and telecommunication networks.
THE ROLE:
We are seeking a 24-month post-doctoral fellow to bridge theoretical
analysis and computational implementation.
Key responsibilities will be drawn from the following:
- Develop new network-adapted moment-SOS hierarchies that exploit sparsity.
- Theoretically analyze these new hierarchies for relaxation gaps.
- Implement space-time decomposition methods (inspired by FEM) to reduce
problem size.
- Improve numerical conditioning using interpolation bases and exploit
low-rank solution structure.
EXPECTED PROFILE:
We seek a candidate with a PhD in Applied Mathematics, Control Theory,
Optimization, or a related field.
The position is flexible and can be tailored to the profiles:
- Analysis-Oriented: Strong background in functional analysis, measure
theory, convex optimization,
and (ideally) PDE theory.
- Computation-Oriented: Strong background in numerical methods for
optimization (SDP), numerical linear algebra,
and scientific computing (Julia, Matlab, or Python).
WE OFFER:
- A 24-month funded researcher position at LAAS-CNRS, Toulouse, France.
- A dynamic, international research environment.
- Travel budget for conferences and multiple research visits to FAU,
Erlangen, Germany.
HOW TO APPLY:
Please send your CV, a cover letter, and contact info for 2-3 references to:
Didier Henrion <henrion@laas.fr>