Speaker: Wenxin Zhang, PhD Candidate in Biostatistics, UC Berkeley
Register HERE
Abstract: Modern decision-making is increasingly operationalized through adaptive decision systems that sequentially learn from data and adapt decisions through feedback loops, including randomized trials with adaptive designs and digital intervention platforms that dynamically assign personalized treatments to improve outcomes and enhance statistical efficiency. However, such adaptive systems raise statistical challenges in how information is used to guide decisions, how decisions are updated over time, and how to make inference from the adaptively-collected data. In this talk, I will present work that develops causal inference and adaptive experimentation methods grounded in the targeted learning framework for designing, evaluating, and learning from adaptive decision-making systems—enabling effective information use, principled decision strategies, and valid inference, with applications to precision health and regulatory science.
The Department of Epidemiology & Biostatistics welcomes all participants to our events. If you need a reasonable accommodation to participate in this event because of a disability, please contact Liz Buggs ([email protected]).