Franck Djeumou
Papers
2
Total Citations
24
H-Index
2
About
Franck Djeumou is an emerging researcher at the intersection of machine learning, control theory, and dynamical systems modeling, with a particular focus on physics-informed neural networks and uncertainty-aware learning frameworks. His work addresses one of the central challenges in modern AI: how to build data-efficient models that respect the fundamental physical laws governing real-world systems. Djeumou's most recognized contribution, "Neural Networks with Physics-Informed Architectures and Constraints for Dynamical Systems Modeling" (2021, 22 citations), demonstrates how embedding a priori physical knowledge — such as conservation laws and structural properties like Jacobian constraints — into deep neural network architectures can dramatically improve generalization and reduce data requirements. Building on this foundation, his 2023 work on physics-constrained neural stochastic differential equations pushes the frontier further, presenting algorithms capable of learning reliable controlled dynamics models from as little as three minutes of observed data, while rigorously quantifying uncertainty. These contributions are especially significant for robotics, autonomous systems, and scientific computing, where data collection is expensive and safety-critical. Djeumou's research signals a broader vision: making learned models not only powerful but principled, interpretable, and trustworthy for deployment in complex real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2