With the Method Development Unit (MDU), MATH+ enhances synergy in method development across the AAs and EFs. The goal is to centralize method development, in particular in modeling, simulation, optimization, and AI, while ensuring it is informed by and closely aligned with the needs and goals of AAs and EFs. Projects within MDU contribute directly to an AA or EF, fostering a more integrated research ecosystem. The MDU encourages open, cross-disciplinary collaboration among researchers in various AAs and EFs, sharing knowledge and applying methods in practical contexts.
Scientist in Charge: Michael Joswig
- MDU-1: Word Equations with Constraints and String Solvers
Laura Ciobanu Radomirovic - MDU-2: Advanced Partitioning Methods for High-Performance Agent-Based Simulations
Sarah Wolf, Tim Conrad, Thomas Steinke - MDU-3: Scaling Limits of Deep Neural Networks: Rough Analysis Meets Optimization
Peter K. Friz, Christian Bayer, Sebastian Kassing - MDU-4: Efficient Multi-Parameter Persistent Homology
Fabian Lenzen
- MDU-1: Word Equations with Constraints and String Solvers