Talk Title: Classical Statistics in the Age of AI

Abstract: Researchers increasingly use generative AI to create “digital twins” and conduct synthetic experiments. Recognizing that LLMs often fail to capture complex real-world behavior, a growing literature has proposed novel methods for combining synthetic and ground-truth data. This talk illustrates the continued relevance of classical statistics for this challenge through two projects. In the first project, we argue that off-the-shelf linear regression is a natural approach for incorporating AI predictions into experiments, building on the randomization inference framework dating back to Fisher and Neyman. Unlike many recent proposals, standard linear regression inherits a “do no harm” property in that the adjusted estimator automatically reverts to the unadjusted difference in means when AI predictions are uninformative. In the second project, we examine how to improve the AI predictions themselves using “activation steering,” a technique from mechanistic interpretability that modifies internal LLM activations to shift behavior toward a target concept. We show that many steering methods implicitly estimate the average gradient of an outcome regression, a quantity with a long history in causal inference and econometrics. This connection immediately enables more flexible models, including Neyman-orthogonal estimators, which in turn lead to improved AI predictions for the first project’s regression framework. Together, these results show how classical statistical ideas continue to provide both conceptual clarity and empirical gains at the intersection of causal inference and generative AI.

Speaker: Avi Feller, Associate Professor in the Goldman School of Public Policy and the Department of Statistics, UC Berkeley

Speaker Bio: Avi Feller is an associate professor in the Goldman School of Public Policy and the Department of Statistics at UC Berkeley, working at the interface of data science and the social sciences. His research focuses on developing practical, transparent methods that can be applied at scale, and on deploying these tools in a range of policy domains and industry applications. His research has appeared in top methodological journals (such as Journal of the American Statistical Association, Journal of the Royal Statistical Society, and Econometrica), in leading machine learning conferences (such as NeurIPS and ICML), and in interdisciplinary journals like the Proceedings of the National Academy of Sciences.

Outside of academia, Feller is a research consultant at Adobe and was previously a visiting researcher at Google. He also co-founded EveryDay Labs, an edtech company focused on reducing student absenteeism (acquired in 2026).

Feller has received multiple awards for his work, including the COPSS Emerging Leader Award, the SREE Early Career Award, the American Statistical Association’s Outstanding Statistical Application Award, and the Mid-Career Award from the ASA Social Statistics Section. He received a PhD in statistics from Harvard University, MSc in Applied Statistics from Oxford University as a Rhodes Scholar, and BA from Yale University. Prior to his doctoral studies, he worked as a policy official in the White House Office of Management and Budget.

The logistics details for each session will be provided to registered participants.  Sign up here to be included on the mailing list and learn about future events.

This talk is co-sponsored by USF's Master's in Applied Economics and the Stanford Causal Science Center. For additional information and abstracts from past talks, please click here.