Research

I am largely interested in dynamical problems, e.g., designing (optimization) algorithms and control systems. In particular, I like to understand the interplay between the underlying problem structure and achievable qualitative behaviour.

Selected (pre)prints

empty 

On topological properties of compact attractors on Hausdorff spaces
Wouter Jongeneel
European Control Conference (ECC), pp. 186 - 191, 2024.
|pdf| |arXiv| |link (IEEE)| |bibtex|

empty 

A Large Deviations Perspective on Policy Gradient Algorithms
Wouter Jongeneel, Daniel Kuhn and Mengmeng Li (alphabetic)
Proceedings of the 6th Annual Learning for Dynamics & Control Conference (L4DC), PMLR vol. 242, pp. 916 - 928, 2024.
|arXiv| |link (PMLR)| |bibtex|

empty 

Topological Obstructions to Stability and Stabilization: History, Recent Advances and Open Problems
Wouter Jongeneel and Emmanuel Moulay
SpringerBriefs in Control, Automation and Robotics (Open Access), pp. X, 132, 2023.
|pdf| |link (Springer)| |Supported by the SNSF|

ZO 

On continuation and convex Lyapunov functions
Wouter Jongeneel and Roland Schwan
IEEE Transactions on Automatic Control (TAC), in press, pp. 1 - 12, 2024.
|pdf| |arXiv| |link (IEEE)| |bibtex| |Bifurcation GIF|

ZO 

Imaginary Zeroth-Order Optimization
Wouter Jongeneel
submitted 2022.
|pdf| |arXiv| |bibtex| |PDE GIF|
As presented at the International Conference on Continuous Optimization 2022

NCost 

Small errors in random zeroth-order optimization are imaginary
Wouter Jongeneel, Man-Chung Yue and Daniel Kuhn
SIAM Journal on Optimization (SIOPT), vol. 34, no. 3, pp. 2638 - 2670, 2024.
|pdf| |arXiv| |link (SIAM)| |bibtex| |Lorenz GIF|
As presented at the SIAM Conference on Optimization 2021
*Winner of the 2023 INFORMS Optimization Society Student Paper Prize |slides|

MDPdef 

Topological Linear System Identification via Moderate Deviations Theory
Wouter Jongeneel, Tobias Sutter and Daniel Kuhn
IEEE Control Systems Letters, vol. 6, pp. 307-312, 2022.
|arXiv| |link (IEEE)| |bibtex|
Presented at the 2021 CDC

eig 

Efficient Learning of a Linear Dynamical System with Stability Guarantees
Wouter Jongeneel, Tobias Sutter and Daniel Kuhn
IEEE Transactions on Automatic Control (TAC), vol. 68, no. 5, pp. 2790 - 2804, 2023.
|arXiv| |link (IEEE)| |bibtex|
As presented at the SIAM Conference on Optimization 2023 |slides|
*Runner up IEEE CSS Swiss Chapter Young Author Best Journal Paper Award 2023

SP 

On Topological Equivalence in Linear Quadratic Optimal Control
Wouter Jongeneel and Daniel Kuhn
European Control Conference (ECC), pp. 2002 - 2007, 2021.
|preprint pdf| |link (IEEE)| |bibtex|

RLQR 

Robust Linear Quadratic Regulator: Exact Tractable Reformulation
Wouter Jongeneel, Tyler Summers and Peyman Mohajerin Esfahani
IEEE Conference on Decision and Control (CDC), pp. 6742 - 6747, 2019.
|pdf| |link (IEEE)| |bibtex|

With special thanks to Daniel Kuhn, Mengmeng Li, Peyman Mohajerin Esfahani, Emmanuel Moulay, Roland Schwan, Tyler Summers, Tobias Sutter and Man-Chung Yue.

Throughout, kindly supported by the EU, l'EPFL, the TU Delft, the NCCR Automation, the SNSF and l'UCLouvain.

Selected talks (feel free to request slides or references, I aim to put everything (relevant) online)

  • A Large Deviations Perspective on Policy Gradient Algorithms,
    July 2024, University of Konstanz, Department of Computer and Information Science.

  • Fika in Paris: on qualitative reasoning in a quantitative age,
    May 2024, KTH Stockholm, Division of Decision and Control Systems (DCS).

  • Control theory with Heinz Hopf,
    February 2023, ETH Zürich, Institut für Automatik (IfA).

  • Topological obstructions to stability and stabilization,
    March 2022, University of Warsaw, Department of Mathematical Methods in Physics, Theory of Duality seminar.

  • From Correlated Data to Guarantees in Stable Identification,
    October 2020, ETH Zürich, Institut für Automatik (IfA).

Other

Stability via reverse I-projections, a symplectic perspective on the computation
Wouter Jongeneel
technical note, 2022
|pdf|

From Moderate Deviations Theory to Distributionally Robust Optimization: Learning from Correlated Data
Tobias Sutter, Wouter Jongeneel and Daniel Kuhn
|video at SPS|

Thesis

Stability: a search for structure
Wouter Jongeneel
PhD Thesis, Electrical Engineering, EPFL 2024, adviser Daniel Kuhn.
|link|

Controlling the Unknown: A Game Theoretic Perspective
Wouter Jongeneel
MSc Thesis, Systems & Control, TU Delft 2019, adviser Peyman Mohajerin Esfahani.
|pdf| |link| |bibtex|