Reinforcement Learning

Reinforcement learning is an area of machine learning that has garnered a lot of attention in recent years. The principle of reinforcement learning is based on the interaction between a decision-making agent and its environment, where the agent learns to perform a task by interacting with the environment while trying to maximize a reward function.

The research goal of the UiT Machine Learning Group has been to employ reinforcement learning to develop automated control systems for artificial pancreas' on order to provide automatic type 1 diabetes management. In addition to the type 1 diabetes controller challenge, our research has fostered several side tracks, like reinforcement learning for Climatology applications and Feature representation for reinforcement learning using functional data analysis.

Highlighted Publications

1. Controlling Blood Glucose Levels in Patients With Type 1 Diabetes Using Fitted Q-Iterations and Functional Features (2018 IEEE 28th International Workshop on Machine Learning for Signal Processing)

Jonas Nordhaug Myhre, Ilkka K. Launonen, Susan Wei and Fred Godtliebsen

2. Reinforcement-learning optimal control for type-1 diabetes (2018 IEEE EMBS International Conference on Biomedical & Health Informatics)

Phuong D. Ngo, Susan Wei, Anna Holubová, Jan Muzik and Fred Godtliebsen