Zach Shahn is currently a researcher in the Health Analytics group of IBM Research. Before that, he was a postdoctoral fellow at the departments of Epidemiology and Biostatistics at Harvard. Previously, he completed a PhD at Columbia University with David Madigan as his advisor. Zach's main research focus is causal inference, i.e. “What causal conclusions can be drawn from data? What methods can be used to draw those conclusions? And what assumptions are necessary?” He is particularly interested in learning optimal dynamic treatment strategies from observational longitudinal healthcare data, e.g. optimal strategies for administering vasopressors to sepsis patients in the intensive care unit. In class, he will likely stress too much the dangers of applying statistical algorithms without first carefully defining the underlying substantive questions of interest.