Outcome Reasoning: the under-discussed engine powering black box development

                                                                       

MIKE BAIOCCHI – STANFORD UNIVERSITY                          JORDAN RODU – UNIVERSITY OF VIRGINIA

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 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, but 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 we will discuss its current use (i.e., as “the common task framework”) and its limitations. We will show why we find a large class of prediction-problems are inappropriate for this new type of reasoning. We will then discuss a way to extend this type of 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.

RELATED PAPERS:

• https://muse.jhu.edu/article/883478
• https://muse.jhu.edu/article/799741