Talk Title: Learning Causes and Using Them

The collection of massive observational datasets has led to unprecedented opportunities for causal inference, such as using electronic health records to identify risk factors for disease. New causal inference methods enable us to learn highly complex models from these datasets, but what happens when people attempt to use them? In this talk, Professor Steinberg will discuss new methods that allow causal relationships to be reliably inferred from complex observational data, work on understanding how people use these models for decision-making, and what it means for inferring causal models that are useful and usable.