Sim2Science: ML with Imperfect Scientific Models

Sim2Science examines what happens when learning methods depend on imperfect scientific simulators. Researchers connect applications across natural sciences to study how modelling errors affect conclusions and decisions, seeking ways to assess reliability beyond the performance of the learning component alone. Session: 08:30–18:15 (Europe/Paris). A paid NeurIPS virtual pass includes workshop livestreams; on-site access follows the host-location registration rules.