Wouter Kouw

Wouter Kouw

Assistant Professor

TU Eindhoven

Biography

Wouter Kouw is a researcher in the Bayesian Intelligent Autonomous Systems lab of TU Eindhoven. He has a dual background in neuroscience and computer science, and is most interested in neuro-inspired forms of artificial intelligence. His research focuses on analyzing, designing and developing probabilistic machine learning systems based on models of information processing in the brain. Specifically, the development of variational Bayesian inference algorithms implemented as message passing on graphs. His work has been applied to nonlinear system identification, adaptive control systems and mobile robotics.

Interests
  • Machine Learning
  • Bayesian Inference
  • Intelligent Systems
  • Neural Computation
Education

  • PhD in Computer Science, 2018

    Delft University of Technology

  • MSc in Neuroscience, 2013

    Maastricht University

Featured Publication

Wouter Kouw
December 2023 IEEE Control Systems Letters
Information-seeking polynomial NARX model-predictive control through expected free energy minimization
Information-seeking polynomial NARX model-predictive control through expected free energy minimization

An adaptive model-predictive controller that balances driving the system to a goal state and seeking system observations that are informative with respect to the parameters of a nonlinear autoregressive exogenous model.

Publications

Tim Nisslbeck, Wouter Kouw.
Coupled autoregressive active inference agents for control of multi-joint dynamical systems.
International Workshop on Active Inference, 2024.
Code Project
Sepideh Adamiat, Wouter Kouw, Bart van Erp, Bert de Vries.
Message passing-based Bayesian control of a Cart-Pole system.
International Workshop on Active Inference, 2024.
Code Project
Wouter Kouw.
Planning to avoid ambiguous states through Gaussian approximations to non-linear sensors in active inference agents.
International Workshop on Active Inference, 2024.
PDF Cite Code
See all publications

Projects

*
All Active Inference Walking robots
ABIAS
Analysis of Bayesian Intelligent Autonomous Systems focuses on the information-theoretic basis of Active Inference. What are the best data points for an agents? How optimal is data collection based on minimizing expected free energy?
ABIAS
FEPQuad
Design of an intelligent autonomous system for quadrupedal robot locomotion using Active Inference (AIF). AIF is a neuroscience-based framework of perception and action, and could bring us closer to how animals learn to walk.
FEPQuad
BayesBrain
Computation in biological brain tissue consumes several orders of magnitude less power than silicon-based systems. Motivated by this fact, this project aims to develop the world’s first hybrid neuro-in-silico Artificial Intelligence (AI) computer.
BayesBrain
SeqDAIS
Sequential Domain Adaptive Intelligent Systems focuses on ordered domain-adaptative classification over a sequence of biased samples from an underlying population.
SeqDAIS

Talks

Embodied artificial intelligence through free energy minimization
Artificial Intelligence should automate dangerous jobs instead of creative ones. I will present a framework for designing intelligent autonomous systems based on the Free Energy Principle, a leading physics-based theory of information processing in brains.
19 Sep 2024 16:20 — 16:30 TU Eindhoven, Netherlands
Planning to avoid ambiguous states through Gaussian approximations to non-linear sensors
Gaussian approximations to nonlinear observation functions hat are sensitive to curvature, such as a second-order Taylor approximation, produce a state-dependent ambiguity term in expected free energy minimization.
11 Sep 2024 10:40 — 14:50 Corpus Christi College, Oxford, United Kingdom
Bayesian grey-box identification of nonlinear convection effects in heat transfer dynamics
We propose a computational procedure for identifying convection in heat transfer dynamics based on a Gaussian process latent force model.
22 Aug 2024 10:30 — 11:00 Northumbria University, Newcastle-upon-Tyne, United Kingdom
See all events

Teaching

TU/e 5XIF0
Neuro Computation: a BSc course covering the computational properties of neurobiological systems and how to design spiking neural networks using mixed-signal “neuromorphic” electronic circuits.
Last updated on 28 Jun 2023
TU/e 5XIF0
TU/e 5XSL0
Fundamentals of Machine Learning: a BSc course that covers the basics of machine learning, including linear classification and regression, support vector machines, random forests, and neural networks.
Last updated on 4 Jan 2022
TU/e 5XSL0
TU/e 5SSD0
Bayesian Machine Learning & Information Processing: a MSc course that covers the fundamentals of the Bayesian (i.e., probabilistic) approach to machine learning and information processing systems.
Last updated on 4 Jan 2022
TU/e 5SSD0

Contact