Reinforcement learning for healthcare

Image: Luke Chesser

Reinforcement learning is a branch of artificial intelligence that provides methods to optimize sequential decisions for long-term outcomes. It has many potential applications in healthcare, such as improving treatment strategies, enhancing patient safety, and reducing costs.

Possible topics

  • Contextual Bayesian non-stationary treatment recommendations
  • Data safe havens, federated learning and other privacy-enhancing techniques
  • Improving human board game performance through reinforcement learning
  • Reinforcement learning for complex adaptive individual health interventions
  • Simulation of patient data for micro-randomized trials

Further reading

  • Shrestha, S, Jain, S. A Bayesian-bandit adaptive design for N-of-1 clinical trials. Statistics in Medicine. 2021; 40: 1825– 1844.
Sebastian Vollmer
Sebastian Vollmer
Professor for Applications of Machine Learning

My research interests lie at the interface of applied probability, statistical inference and machine learning.