Since late May, Álvaro Romero Castillo, a third-year undergraduate student in Theoretical Physics at Trinity College Dublin, has been conducting research at the Cluster of Excellence SimTech, University of Stuttgart. He joined the Junior Research Group of Miriam Klopotek through the DAAD RISE Germany program and will stay until August 17.
Álvaro contributes to ongoing work at the intersection of many-body simulations and machine learning, a core focus of Klopotek’s research group. The group investigates non-equilibrium many-body dynamics using statistical-physical model systems and explores how data-driven methods such as self-supervised learning, generative models, and spatio-temporal prediction can be integrated into simulation-based science. A particular interest lies in interpreting machine learning models using physical insight and in developing physics-informed learning methods for complex systems.
During his internship, Álvaro is working closely with doctoral researcher Mario Gaimann on the project “Reservoir Computing with Active Matter.” The central question: under what physical conditions can a swarm-like system - a collection of interacting active particles - be used as a computational substrate for machine learning tasks?
The project Álvaro is contributing to recent work from the Klopotek group, where swarms of active particles are investigated as computational reservoirs - physical systems that process information through their inherent dynamics. In particular, the research explores how memory and nonlinearity, two key properties for machine learning, emerge in such systems under different physical conditions, such as particle density, noise, or interaction strength. The study introduces minimal models of interacting active particles and evaluates their computational power using standard machine learning benchmarks - laying the foundation for using physical matter as an unconventional computing substrate.
The internship offers Álvaro a first immersion into collaborative research and a highly interdisciplinary setting that connects statistical physics, dynamical systems, and AI methods. With prior experience in high-energy physics data analysis and a strong interest in the mathematical structures underlying physical systems, Álvaro brings both technical skills and scientific curiosity to the project.
“This is my first experience in a collaborative scientific research environment,” Álvaro says. “Working in a group that bridges machine learning and theoretical physics - and doing so in Stuttgart, at the heart of Cyber Valley - makes this a truly inspiring opportunity.”
His placement at SimTech was coordinated by doctoral researcher Patrick Egenlauf, who is currently abroad on a SimTech Stay Abroad in Vienna. Mario Gaimann has taken over the day-to-day supervision during this period. This handover exemplifies SimTech’s cooperative research culture and its commitment to early-career mentoring and international knowledge exchange.
The DAAD RISE Germany program offers research internships to students from North America, the UK, and Ireland. This year, over 2,500 applications competed for approximately 250 positions. Hosting one of these internships underscores SimTech’s role as an international hub for interdisciplinary research and talent development.
Learn more about the DAAD RISE Germany program