Nilo Schwencke

Nilo Schwencke

Postdoc @ LIP, ENS Lyon on PINNs & SciML

Nilo Schwencke

I am a Postdoctoral Researcher at the OCKHAM Team (INRIA; LIP, ENS Lyon; CNRS), under the supervision of Elisa Riccietti and Nelly Pustelnik. My research revolves around Scientific Machine Learning (SciML), numerical analysis, and kernel methods.

I completed my PhD in December 2025 at Université Paris-Saclay, under the supervision of Cyril Furtlehner (TAU Team, INRIA Saclay – A&O–LISN–CNRS). My dissertation developed improved training schemes for PINNs combining natural gradients and kernel methods.

My research interests are:

  • Physics-Informed Neural Networks (PINNs) — algorithmic design and mathematical foundations
  • Natural gradient methods
  • Connections between neural network-based solvers and classical numerical methods (Galerkin methods, FEM)
  • Reproducing Kernel Hilbert Spaces (RKHS)

Please refer to my publications and talks pages for a list of my recent research activities, and to upcoming talks for future venues. I am also a contributor to SciMBA, an open-source SciML library implementing PINNs, neural Galerkin schemes and natural gradient techniques for solving parametric PDEs.

Short bio