Hands on Machine Learning for Fluid Dynamics 2026
Date: 02 - 06 February 2026
Location: online / on-site: Waterloosesteenweg, 72, 1640 Sint-Genesius-Rode, Belgium - near Brussels
About the course
Now in its sixth edition, the Hands-on Machine Learning for Fluid Dynamics course has become a leading international training program, with more than 500 participants having attended its previous editions from academia, research institutes, and industry. This one-week intensive course provides a practical and accessible introduction to modern machine learning techniques, tailored to the challenges and opportunities of fluid mechanics.
The course combines concise theoretical lectures with hands-on Python sessions, guiding participants to frame and solve problems in fluid dynamics using data-driven approaches. Topics include regression and uncertainty quantification, dimensionality reduction, reduced-order modeling, turbulence modeling, system identification, data assimilation and inverse problems for digital twinning, as well as reinforcement learning for flow control. Each day blends expert instruction with interactive exercises designed to build intuition and connect algorithms to physical understanding.
The program welcomes participants with diverse backgrounds—from graduate students entering the field to researchers and professionals seeking to expand their toolkit. The course begins gently, establishing key concepts and intuition, but progressively accelerates toward advanced, state-of-the-art topics. These deeper segments can be followed at different levels of detail: participants may grasp the main ideas during the lectures or explore them further afterward through the extensive set of exercises and tutorials provided. This structure ensures that both newcomers and experienced attendees benefit fully, gaining both conceptual understanding and a strong foundation for continued self-study.
Programme
Monday, 2 February 2026: Introduction and Fundamentals of Optimization
08:30 - 09:00 Registration and Welcome Coffee
09:00 - 09:15 Welcome Address and Course Overview
Miguel A. Mendez, von Karman Institute
09:15 - 10:45 Lecture 1: What is Machine Learning?
Miguel A. Mendez, von Karman Institute
10:45 - 11:15 Coffee Break
11:15 - 12:45 Lecture 2: Regression and Uncertainty Quantification — Hands-on Session
Miguel A. Mendez, von Karman Institute
12:45 - 14:00 Lunch Break
14:00 - 15:30 Lecture 3: Gradient-Based Optimization — Principles and Tools
Romain Poletti, von Karman Institute
15:30 - 16:00 Coffee Break
16:00 - 17:30 Lecture 4: Bio-Inspired Optimization — From Genetic Algorithms to Particle Swarms
Miguel A. Mendez, von Karman Institute
17:30 End of Day
Tuesday, 3 February 2026: From Bases to Kernels and Neural Networks: Modern Regression Method
09:00 - 10:30 Lecture 5: Regularized and Constrained Radial Basis Function Models
Manuel Ratz, von Karman Institute
10:30 - 11:00 Coffee Break
11:00 - 12:30 Lecture 6: Bayesian Formalism and Kernel-Based Learning
Miguel A. Mendez, von Karman Institute
12:30 - 14:00 Lunch Break
14:00 - 15:30 Lecture 7: Gaussian Processes for Multifidelity Learning and Bayesian Optimization
Miguel A. Mendez, von Karman Institute
15:30 - 16:00 Coffee Break
16:00 - 17:30 Lecture 8: Deep Learning and Physics-Informed Neural Networks
Miguel A. Mendez, von Karman Institute
17:30 End of Day
Wednesday, 4 February 2026: Learning and Data Assimilation in Dynamical Systems
09:00 - 10:30 Lecture 9: Machine Learning for Turbulence Modeling
Matilde Fiore, von Karman Institute
10:30 - 11:00 Coffee Break
11:00 - 12:30 Lecture 10: Propagating Gradients through Dynamical Systems with the Adjoint Method
Miguel A. Mendez, von Karman Institute
12:30 - 14:00 Lunch Break
14:00 - 15:30 Lecture 11: Learning Dynamical Systems and Time Series Forecasting
Samuel Ahizi, von Karman Institute
15:30 - 16:00 Coffee Break
16:00 - 17;30 Lecture 12: Statistical Identification and Data Assimilation with Kalman Filter(s)
Miguel A. Mendez, von Karman Institute
17:30 End of Day
Thursday, 5 February 2026: Dimensionality Reduction and Latent Space Modeling
09:00 - 10:30 Lecture 13: Linear Autoencoders and Data-Driven Modal Decomposition
Miguel A. Mendez, von Karman Institute
10:30 - 11:00 Coffee Break
11:00 - 12:30 Lecture 14: Manifold Learning and Nonlinear Dimensionality Reduction
Miguel A. Mendez, von Karman Institute
12:30 - 14:00 Lunch Break
14:00 - 15:30 Lecture 15: Variational Autoencoders — Probabilistic Representation Learning
Joachim Dominique, Cenaero
15:30 - 16:00 Coffee Break
16:00 - 17:30 Lecture 16: From Spectral Methods to Data-Driven Reduced-Order Models
Miguel A. Mendez, von Karman Institute
17:30 End of Day
Friday, 6 February 2026: Flow Control, Reinforcement Learning, and Digital Twinning
09:00 - 10:30 Lecture 17: From Optimal Control to Reinforcement Learning — Fundamental Concepts
Miguel A. Mendez, von Karman Institute
10:30 - 11:00 Coffee Break
11:00 - 12:30 Lecture 18: Value-Based and Policy-Based Reinforcement Learning
Lorenzo Schena, von Karman Institute
12:30 - 14:00 Lunch Break
14:00 - 15:30 Lecture 19: The Actor–Critic Formalism in Reinforcement Learning
Romain Poletti, von Karman Institute
15:30 - 16:00 Coffee Break
16:00 Lecture 20: Reinforcement Twinning — From Digital Twins to Model-Based Control
Miguel A. Mendez, von Karman Institute
17:30 End of Day
Fee and Registration
- Deadline for on-site registration: 19 January 2026
- Deadline for online registration: 26 January 2026
- Discount of 10% in case of online participation*
- Discount of 30% for citizens from NATO countries*
* applicable on the price listed below
Early Bird Registration until 2 December 2025
- Undergraduate students: €315
- PhD students: €864
- Staff from recognized universities / research center: €1368
- Staff from commercial organizations: €1728
Late Registration after 2 December 2025
- Undergraduate students: €350
- PhD students: €960
- Staff from recognized universities / research center: €1520
- Staff from commercial organizations: €1920