Talk Abstract: Causal decision making (CDM) at scale has become a routine part of business, and increasingly CDM is based on statistical models and machine learning (ML) algorithms. Businesses target offers, incentives, recommendations, and even content algorithmically with the goal of affecting consumer behavior. I highlight something important: targeting for causal effect is not the same as causal effect estimation. In fact, accurate causal effect estimation is not necessary for accurate CDM. I will discuss three implications: (1) We should optimize ML for accurate causal targeting rather than for accurate effect estimation. (2) For CDM, it may be just as good or even better to learn with confounded data as with unconfounded data. Finally, (3) causal statistical modeling may not be necessary at all to support CDM, because there may be (and perhaps often is) a proxy target for statistical modeling that can do as well or better. This observation helps to explain at least one broad common CDM practice that seems ``wrong" at first blush--the widespread use of non-causal models for targeting interventions like advertisements and retention incentives. The last two implications are particularly important in practice, as acquiring (unconfounded) data on both ``sides" of the counterfactual for modeling can be quite costly and often is impracticable. Understanding causal targeting is vital to modern data science practice, and is fertile ground for new data science research (there's been surprisingly little until just the past few years).
Sennott Square Building, Room 5317
Zoom (password: CS2022) Virtual attendees must be signed into Zoom with their university account or chose "Sign in with SSO" to attend this meeting.
Note: CS2003 students attending in person must sign in and out of the event.
Biosketch: Foster Provost is Ira Rennert Professor of Entrepreneurship, Data Science, and Information Systems and Director, Fubon Center, Data Analytics & AI, at the Stern School of Business at New York University. He is also a Distinguished Scientist for real-estate giant Compass. Foster received his PhD from Pitt's CS Department in 1993. He previously was Editor-in-Chief of the journal Machine Learning. Foster's research on data science and machine learning has won many awards, including (among others) the 2020 ACM SIGKDD Test of Time Award, the 2017 European Research Paper of the Year, Best Paper awards in the top data science research venues across four decades, the 2009 INFORMS Design Science Award, IBM Faculty Awards, and a President’s Award from NYNEX Science and Technology (now Verizon). His book Data Science for Business is a perennial best seller and was listed as one of Fortune Magazine's "must read books for MBAs." Foster has helped to found several successful startups, including Integral Ad Science (which IPO'ed last year), adtech powerhouse Dstillery, and Detectica (acquired by Compass, which also IPO'ed last year). Foster's Progressive Rock album Mean Reversion is available in all the usual places. He will deliver a CS Colloquium series talk on Friday, February 25th at 2 p.m.
Host: Daniel Mosse