Published:
This project proposes to build a system based on physics-informed GNN to the problem of multi-pollutant air quality forecasting across the Netherlands. Read more
Published:
This project aims to design RL-based control strategies for safe insulin dosing for Type 1 Diabetes (T1D) patients. Read more
Published:
This project investigates the integration of first-principle fluid dynamic models into deep learning frameworks to enhance air pollution time series forecasting. Read more
Published:
If you have a problem or idea, we may consider it for your bachelor’s or master’s project. Read more
Published:
In this project, we aim to learn Lyapunov functions using DNNs and leverage them to guide the training of actor-critic agents toward safe behavior. Read more
Published:
We aim to explore and evaluate the capabilities of a hybrid system known as KalmanNet. Read more
Published:
This project explores how well multi-agent reinforcement learning works in simulated football games and how scalable it is. Read more
Published:
This project explores the advantages and limitations of using Gaussian Process in model-free Reinforcement Learning , focusing on Q-learning. Read more
Published:
Design a data-driven planning system for fish catching routes based on historical data; Read more
Published:
This project aims to develop a data preprocessing pipeline and use deep learning models to predict grid quality and resilience in a simulated neighborhood with distributed energy resources. Read more
Published:
This project aims to develop tools to teach machine learning concepts with hands-on approach focusing on teenagers above 16 years old. The resulting tools are expected to catch and maintain the motivation of the students while teaching abstract concepts in an easy-to-follow way. Read more
Published:
This project proposes to develop and evaluate a physics-enhanced deep learning approach for smog cloud modelling by using the growing amount of data on the topic, new deep learning architectures for time-series processing, and the current models as heuristics. Read more
Published:
This project aims to benchmark safe RL and propose novel ideas to improve the existing state-of-the-art in the area. Read more