Cognitive science · Computer science
Hannes P. Voß
Exploring cognitive processes in the brain, whole-brain modeling, and machine learning for Type 1 Diabetes.
About me
Welcome! I am interested in cognitive processes in the human brain — and especially in replicating these processes in cognitive systems.
I am also interested in whole-brain modeling of connectomes, particularly bio-inspired connectomes: transferring empirically collected structural connectomes into neural networks, enabling continuous learning in these networks, and analysing them.
Beyond that, I am strongly interested in Type 1 Diabetes and in applying machine learning to diabetes-related challenges such as unannounced meal detection and automated insulin delivery. Previously, I worked as a health app developer on both private and commercial projects.
On this site you can find my publications, project write-ups and blog posts. If you are interested in working with me, feel free to send me an email.
Research interests
Cognitive systems
Understanding cognitive processes in the human brain and replicating them in artificial cognitive systems.
Connectome modeling
Whole-brain models built from empirical structural connectomes, and continual learning in these bio-inspired networks.
ML for Type 1 Diabetes
Deep learning for unannounced meal detection and deep reinforcement learning for automated insulin delivery.
Writing
All posts →-
Controlling Insulin Delivery with Deep Reinforcement Learning
A deep reinforcement learning approach — Deep Q-Learning, to be precise — for controlling insulin delivery in people with Type 1 Diabetes (T1D).
-
In progress
Unannounced Meal Detection using Long Short-Term Memory Networks
A deep learning approach to detect unannounced meals for people with Type 1 Diabetes (T1D).
-
In progress
Sequence Recall in Reservoir Computing using Echo State Networks
How to solve sequence recall tasks in reservoir computing using echo state networks.
Projects
All projects →Android app
Magnet
A mobile app for discovering, creating and sharing the best spots for magnet fishing.
Open source
SimGlucose
Open-source Python implementation of the UVa/Padova Type 1 Diabetes simulator, used for reinforcement learning research.
Android app
SmartScale
A mobile app for tracking weight and body measurements.