Gerben Beintema
Department / Institute
RESEARCH PROFILE
Gerben I. Beintema is a post-doc researcher at the Control Systems (CS) Group at the Department of Electrical Engineering. His current research is on data-driven learning of dynamical systems with AI and deep learning, under the supervision of Associate Professor Roland Tóth and assistant Professor Maarten Schoukens. His main research interest is improving and understanding deep nonlinear state-space identification to obtain accurate, robust, and generally applicable methods.
ACADEMIC BACKGROUND
Gerben I. Beintema obtained his PhD in 2024 at the Eindhoven University of Technology (TU/e) under the supervision of prof. Roland Toth and Maarten Schoukens. His research was focused on data-driven learning of dynamical systems using AI and machine learning. He also obtained a master's degree in Applied Physics (Flow profile) from the Eindhoven University of Technology (TU/e), cum laude, in 2019. Under the supervision, of Full Professor Federico Toschi, his MSc thesis showed that deep reinforcement learning can be used to obtain state-of-the-art controllers to reduce fluid movement in a chaotic thermally driven convective flow. This reduction of flow is a desirable property for crystal silicon growth. This research has been successfully published in the peer-reviewed Journal of Turbulence.
Recent Publications
-
G.I. Beintema,Maarten Schoukens,Roland Tóth
Meta-State-Space Learning
Automatica (2024) -
Jonas Weigand,Gerben I. Beintema,Jonas Ulmen,Daniel Görges,Roland Tóth,Maarten Schoukens,Martin Ruskowski
State Derivative Normalization for Continuous-Time Deep Neural Networks
IFAC-PapersOnLine (2024) -
Max Champneys,Gerben I. Beintema,R. Tóth,Maarten Schoukens,Timothy J. Rogers
Baseline Results for Selected Nonlinear System Identification Benchmarks
IFAC-PapersOnLine (2024) -
András Retzler,Roland Tóth,Maarten Schoukens,Gerben I. Beintema,Jonas Weigand,Jean-Philippe Noël,Zsolt Kollár,Jan Swevers
Learning-based augmentation of physics-based models
Data-Centric Engineering (2024) -
Sarvin Moradi,G.I. Beintema,N.O. Jaensson,Roland Tóth,Maarten Schoukens
Output error port-Hamiltonian neural networks
(2024)
Ancillary Activities
No ancillary activities