Continual learning by combining reinforcement learning and data assimilation in the context of precision therapy (B08)

• Subject Area: Mathematics
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 318763901

The variability in the response of patients to drug treatment is one of the key challenges in drug therapy. An important example is cytotoxic chemotherapy. The project aims at a novel class of model-informed precision dosing approaches based on a combination of reinforcement learning and sequential data assimilation that allows for continuous learning across patients, and addresses the key problems of model bias and complexity encountered in real-world scenarios. The aim is to provide decision support to individualise drug treatment based on relevant patient factors combined with therapeutic drug/biomarker monitoring data.

• DFG Programme: Collaborative Research Centres
• Subproject of: SFB 1294: Data Assimilation – The Seamless Integration of Data and Models
• Applicant Institution: Universität Potsdam
• Project Heads: Dr. Niklas Hartung; Professor Dr. Wilhelm Huisinga, since 7/2021; Professor Dr. Manfred Opper, from 7/2021 until 12/2025; Professorin Dr. Jana De Wiljes, since 7/2021