Research
Daniel Musekamp develops data-efficient neural solvers for partial differential equations. His work uses active learning to select informative simulations and investigates how knowledge from large foundation models can be transferred to smaller, less expensive solvers. He also contributes to neural fields and latent-space models for time-dependent physical systems.
Selected publications
- Distillation of Foundation Models for Time-dependent PDEs 2026
- CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs 2025
- Active Learning for Neural PDE Solvers 2025
- Vectorized Conditional Neural Fields: A Framework for Solving Time-dependent Parametric Partial Differential Equations 2024
Information reviewed 26 September 2026 · University profile