Personne

Wouter Jongeneel

Cette personne n’est plus à l’EPFL

Stability: a search for structure

In this thesis we study stability from several viewpoints. After covering the practical importance, the rich history and the ever-growing list of manifestations of stability, we study the following. (i) (Statistical identification of stable dynamical syste ...

EPFL2024

A Large Deviations Perspective on Policy Gradient Algorithms

Daniel Kuhn, Mengmeng Li, Wouter Jongeneel

Motivated by policy gradient methods in the context of reinforcement learning, we derive the first large deviation rate function for the iterates generated by stochastic gradient descent for possibly non-convex objectives satisfying a Polyak-Łojasiewicz co ...

2024

On Continuation and Convex Lyapunov Functions

Roland Schwan, Wouter Jongeneel

Suppose that the origin is globally asymptotically stable under a set of continuous vector fields on Euclidean space and suppose that all those vector fields come equipped with - possibly different - convex Lyapunov functions. We show that this implies the ...

2024

Small Errors in Random Zeroth-Order Optimization are Imaginary

Daniel Kuhn, Wouter Jongeneel

Most zeroth-order optimization algorithms mimic a first-order algorithm but replace the gradient of the objective function with some noisy gradient estimator that can be computed from a small number of function evaluations. This estimator is constructed ra ...

Society for Industrial & Applied Mathematics (SIAM)2024

On topological properties of compact attractors on Hausdorff spaces

Wouter Jongeneel

We characterize when a compact, invariant, asymptotically stable attractor on a locally compact Hausdorff space is a strong deformation retract of its domain of attraction. ...

Institute of Electrical and Electronics Engineers Inc.2024

Efficient Learning of a Linear Dynamical System with Stability Guarantees

Daniel Kuhn, Tobias Sutter, Wouter Jongeneel

We propose a principled method for projecting an arbitrary square matrix to the non- convex set of asymptotically stable matrices. Leveraging ideas from large deviations theory, we show that this projection is optimal in an information-theoretic sense and ...

2023

Topological Linear System Identification via Moderate Deviations Theory

Daniel Kuhn, Tobias Sutter, Wouter Jongeneel

Two dynamical systems are topologically equivalent when their phase-portraits can be morphed into each other by a homeomorphic coordinate transformation on the state space. The induced equivalence classes capture qualitative properties such as stability or ...

2022

On Topological Equivalence in Linear Quadratic Optimal Control

Daniel Kuhn, Wouter Jongeneel

Dynamical systems are topologically equivalent when their orbits can be mapped onto each other via a homeomorphic change of coordinates. We will show that in general, closed-loop systems resulting from Linear Quadratic Optimal Control problems are all topo ...

2021