Professor Michael Baiocchi, Stanford University
Friday April 28, 2023
3:30 pm - 4:05 pm
How to tell the difference between machine learning and traditional (bio)statistics
Abstract:
For a long time in (bio)statistics we only had two fundamental ways of reasoning using data: warranted reasoning (e.g., randomized trials) and model reasoning (e.g., linear regression models). In the 1980s a new, extraordinarily productive way of reasoning about algorithms emerged: "outcome reasoning." Outcome reasoning has come to dominate areas of data science, it has been under-discussed and its impact under-appreciated. For example, it is the primary way we reason about "black box" algorithms. In this talk I'll discuss its current use (i.e., as "the common task framework") and its limitations. I'll show why we find a large class of prediction-problems are inappropriate for this new type of reasoning. I'll then discuss a way to extend outcome reasoning for use, where appropriate, in assessing algorithms for deployment (i.e., when using a predictive algorithm "in the real world"). We purposefully developed this new framework so both technical and non-technical people can discuss and identify key features of their prediction problem (paper: https://muse.jhu.edu/article/883478).