Time: April 11, 4:30-5:30pm
location: SAS 5270
zoom: zoom link (Meeting ID: 989 4520 8567 Passcode: 970176)

Speaker Names: Mike Baiocchi / Associate Professor, Department of Epidemiology and Population Health / Stanford University
Jordan Rodu / Assistant Professor, Department of Statistics / University of Virginia

Pronunciations: bye-oh-key
raw-dew

Title: How to tell the difference between machine learning and (bio)statistics

Abstract: We’ll start this talk discussing a couple of studies: (i) a randomized trial to evaluate a sexual assault prevention program in Nairobi, Kenya and (ii) a remote detection operation to find and disrupt labor trafficking in the Amazon rainforest. These are both “data science” projects but they are wildly different in how they work. What makes them so different? 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.

Main paper: https://muse.jhu.edu/article/883478
Related paper: https://muse.jhu.edu/article/799741

Baiocchi Bio:  Professor Baiocchi is an interventional statistician (i.e., grounded in both the creation and evaluation of interventions). He thinks a lot about behavioral interventions and how to rigorously evaluate if and how they work. Methodologically, he focuses on creating statistically rigorous methods for causal inference that are easy to critique. He designed  — and was the principle investigator for — two large randomized studies of interventions to prevent sexual assault in the settlements of Nairobi, Kenya. He now works a lot on anti-human trafficking with the Brazilian Federal Prosecutors.

Rodu Bio: Professor Rodu’s research interests lie at the intersection of statistics and machine learning. He is interested in both developing principled methodology for incorporating machine learning into the scientific pipeline through a theoretical understanding of how and when it is safe to do so, and leveraging statistical principles to fortify our understanding of machine learning algorithms. In addition, he studies methodology related to high-dimensional time series, with a particular eye towards applications in health and in the environmental sciences. He is also interested in data visualization and developing software to support visualization.

Websites:
https://profiles.stanford.edu/michael-baiocchi
https://jrodu.github.io/