AI & computational efficiency
We develop computational tools that enhance the speed and scalability of infectious disease simulations. As model complexity and data volume grow, efficient algorithms become essential for timely and reliable decision support.
We focus on improving the performance of both population-based and individual-based models. This includes optimizing simulation runtimes, scaling methods to national or regional levels, and integrating AI to automate or accelerate key modeling processes.
In recent work, we explored how reinforcement learning can support epidemic policy design by efficiently navigating complex decision spaces. Our study on COVID-19 mitigation strategies used multi-objective optimization to identify trade-offs between health and socio-economic outcomes (Reymond et al., 2024).
Our technical efforts are supported by collaboration with the Flemish Supercomputing Center (VSC), enabling us to run large-scale simulations and computational experiments efficiently.
We also invest in advanced training. SIMID researchers co-organize the course Modelling and AI for Infectious Disease Control with Pieter Libin (Vrije Universiteit Brussel), covering epidemic modeling, deep reinforcement learning, and policy optimization through practical sessions in Python.
By combining high-performance computing, AI, and epidemiological modeling, we aim to build responsive, scalable tools that meet the demands of modern public health challenges.
