Rembert Daems

I am currently a postdoc at Ghent University - imec, in the group of Thomas Demeester. My main research interests are stochastic differential equations, fractional Brownian motion, diffusion models, and combinations or variations of all these, in the application domain of machine learning for drug discovery.
My PhD dissertation is on the intersection of visual perception, (stochastic) dynamical systems and physics priors. I started (still ongoing) international collaborations with Tolga Birdal and Manfred Opper on SDEs driven by fractional Brownian motion which led to an ICLR spotlight paper. Subsequently, I contributed to work led by Gabriel Nobis on generative fractional diffusion models and fractional Schrödinger bridges.
During my PhD I co-founded Hippo Dx, a medical device start-up. We developed the automated skin prick test device (SPAT), enabling consistent and precise allergen testing and freeing precious medical professionals time. I mainly worked on the camera and custom lighting system, image processing and AI-assisted diagnostics.
Before starting my PhD, I worked in industry for a few years. After graduating from Ghent University in 2016 with a master’s degree in electromechanical engineering I started working at CNH Industrial, in the combine harvester innovation team. There I had the chance to learn a lot about computer vision, and apply deep learning to automate combine harvesters. In 2018 I switched to Octinion, a smaller R&D company, where I worked on many deep learning projects in the agricultural sector. In my role as AI lead I introduced deep learning in the company and coached colleagues on a technical level.
On this website I write about my research as well as side projects. I live in the beautiful city of Bruges with my wonderful wife and three kids.
latest posts
Jun 18, 2025 | Learning from Video in Continuous Time Using Physics Priors and Fractional Noise |
---|---|
Apr 22, 2024 | Variational Inference for SDEs Driven by Fractional Noise |
Jun 10, 2022 | KeyCLD |
selected publications
- KeyCLD: Learning constrained Lagrangian dynamics in keypoint coordinates from imagesNeurocomputing, 2024
- Variational Inference for SDEs Driven by Fractional NoiseIn The Twelfth International Conference on Learning Representations, 2024
- Generative Fractional Diffusion ModelsIn Advances in Neural Information Processing Systems, 2024