INTRAL aims at minimizing the resources required to develop, maintain and adapt data-driven and hybrid (process-) models by developing and applying novel Transfer Learning (TL) approaches. The main focus of the project is the development of interpretable TL methods, which enable the embedding of expert knowledge (interactivity). This will enable application of TL in strictly regulated environments such as the (bio-) pharmaceutical industry and increase the robustness and the reliability of TL models, respectively. The result will be a software-framework that bundles algorithms, workflows and interfaces for the development, interpretation and validation of TL models for applications in the field of process analytical technology (PAT).
