This paper was accepted on the workshop “Machine Studying 4 Bodily Sciences” at NeurIPS 2022.
Hybrid modelling reduces the misspecification of professional bodily fashions with a machine studying (ML) part realized from information. Equally to many ML algorithms, hybrid mannequin efficiency ensures are restricted to the coaching distribution. To deal with this limitation, right here we introduce a hybrid information augmentation technique, termed professional augmentation. Based mostly on a probabilistic formalization of hybrid modelling, we exhibit that professional augmentation improves generalization. We validate the sensible advantages of professional augmentation on a set of simulated and real-world methods described by classical mechanics.