A one-week intensive course built around practical Python sessions for modern machine learning in fluid mechanics. The program blends short theoretical blocks with hands-on exercises on regression, uncertainty quantification, kernels and Gaussian processes, reduced-order modeling, data assimilation for digital twinning, and reinforcement learning for flow control.

A lecture series dedicated to the physics, technologies, and algorithms of passive and active flow control. The program spans fundamentals and key challenges, control algorithms (from optimization to learning), and real-world applications—with a strong focus on data-driven control and reinforcement learning perspectives.

The 2025 edition of the Introduction to Measurement Techniques wrapped up a rich program of lectures, discussions, and demonstrations at the von Karman Institute for Fluid Dynamics. Participants and lecturers engaged deeply with both foundational principles and recent advances in fluid mechanics, fostering lively dialogue and reinforcing this series’ role as a cornerstone of advanced fluid dynamics education.

The first edition of the AI and Fluid Mechanics Conference (AIFluids 2025) brought together an international community at the interface between artificial intelligence and fluid mechanics. Our group participated with six student contributions, covering topics from meshless modeling and data assimilation to digital twinning and flow control.

The conference also featured a keynote lecture on reinforcement twinning, highlighting emerging paradigms for closed-loop learning and real-time decision-making in complex fluid systems.

One more edition of the Hands-on Machine Learning for Fluid Dynamics course has been successfully completed at the von Karman Institute for Fluid Dynamics. This edition expanded the program with new material on digital twinning and hybrid physics–data learning, while maintaining a strong emphasis on practical implementation through intensive Python-based sessions.

The VKI lecture series on Particle Image Velocimetry (PIV), led by Stefano Discetti and Miguel Alfonso Mendez, has successfully concluded after an intense and inspiring week at the von Karman Institute. The program featured 15 lectures by 11 internationally renowned experts, complemented by a laboratory demonstration, a poster session, and many lively discussions on the present and future of PIV.

The third edition of the “Hands-on Machine Learning for Fluid Dynamics” has just been completed. This year, we put more focus on real-time assimilation and digital twinning, introducing the notion of reinforcement twinning (RT). Hundreds of scripts were provided and participants implemented physics-constrained ML, data assimilation, dimensionality reduction, RL and flow control.

After an intense week at LXLASER2024 in Lisbon, we presented several contributions based on SPICY (Super-resolution and Pressure from Image veloCimetrY)—the software we are developing at the von Karman Institute for Fluid Dynamics for meshless data assimilation from image-velocimetry data. Applications covered cryogenic sloshing (Pedro Afonso Marques), turbulent wetting (Damien Rigutto), ensemble statistics in turbulent flows (Manuel Ratz), and meshless POD (Iacopo Tirelli, from UC3M).

We hosted 241 participants and 16 internationally renowned speakers from more than 20 countries for the VKI–ULB lecture series on Machine Learning for Fluid Dynamics.

Another successful edition of the VKI Lecture Series on Fluid Dynamics was held at the von Karman Institute for Fluid Dynamics. Featuring a comprehensive program of lectures, discussions, and demonstrations, the series brought together graduate students and experts to explore both foundational and emerging topics in experimental fluid mechanics.

The second edition of the Hands-on Machine Learning for Fluid Dynamics further strengthened the practical and physics-oriented nature of the course, with an increased focus on meshless integration of PDEs and turbulence modeling using deep learning techniques.

This edition introduced advanced concepts such as Neural ODEs and hybrid physics–data models, alongside intensive Python coding sessions. Participants worked hands-on with supervised, unsupervised, and reinforcement learning approaches applied to modeling, dimensionality reduction and filtering, and flow control problems.

The von Karman Institute lecture series on Introduction to Measurement Techniques has successfully concluded after an intensive week of lectures and practical sessions. The course covered fundamental experimental methods in fluid dynamics, bridging physical principles, measurement uncertainty, and modern diagnostic tools.

As a long-standing VKI classic, this lecture series provides essential training for graduate students, researchers, and engineers seeking a solid foundation in experimental fluid mechanics.

The first edition of the Hands-on Machine Learning for Fluid Dynamics marked the launch of an intensive, practice-oriented lecture series dedicated to modern machine learning methods for fluid mechanics.

Over the course of an intense week, participants explored supervised, unsupervised, and reinforcement learning through guided exercises and tutorials applied to real fluid-dynamics problems, including aeroacoustic noise prediction, turbulence modeling, meshless PDE integration and super-resolution, dimensionality reduction, and flow control.

The course brought together both on-site and online participants and sparked lively technical discussions throughout the week.

A full week at the von Karman Institute for Fluid Dynamics was dedicated to Particle Image Velocimetry (PIV) and Lagrangian Particle Tracking (LPT), covering both fundamental principles and recent methodological advances.

The lecture series was directed by Stefano Discetti, who coordinated an outstanding team of lecturers including Fulvio Scarano, Andrea Ianiro, Tommaso Astarita, Christian Kähler, Andreas Schröder, Andrea Sciacchitano, and Miguel Alfonso Mendez.

The program also featured dedicated laboratory demonstrations, led by Muhsin Can Akkurt and Misuki Okada, offering participants hands-on exposure to state-of-the-art experimental techniques.

The 14th edition of the European Coating Symposium (ECS 2021) was jointly organized by the von Karman Institute for Fluid Dynamics (VKI) and the Université libre de Bruxelles (ULB), and held in a hybrid online–onsite format in Brussels.

In addition to the main symposium, ECS 2021 hosted a dedicated pre-conference training course entitled “Dynamics of Liquid Film Coatings: Theory and Applications”, offering an in-depth overview of the physics, modeling, and industrial relevance of coating flows.

The symposium and the training course were co-organized with Prof. Jean-Marie Buchlin (VKI) and Prof. Benoît Scheid (ULB).

This lecture series marked the first VKI lecture series fully dedicated to machine learning for fluid mechanics. Hosted by the Université libre de Bruxelles (ULB) and held in the classic VKI lecture-series format, the course brought together leading experts at the interface between physics-based modeling and data-driven methods.

Topics spanned data-driven analysis, reduced-order modeling, control, and turbulence closures. The lectures laid the foundation for what would later become a sustained research and education effort on data-driven fluid mechanics at VKI and beyond.

The lecture notes were later expanded into an edited volume published by Cambridge University Press under the title “Data-Driven Fluid Mechanics: Combining First Principles and Machine Learning.”

The series was co-organized with Alessandro Parente (ULB), Andrea Ianiro (UC3M), Bernd R. Noack (HIT Shenzhen / TU Berlin), and Steven L. Brunton (University of Washington).