Inferential Replicability for Online Decision-Making Algorithms
Inferential Replicability for Online Decision-Making Algorithms
-Mathematical Sciences Building
MS 5200
Data Theory Seminar with Susan Murphy
Speaker: Susan Murphy
Time: 11am – 12pm, Nov. 11, 2026
Location: Mathematical Sciences Building, Room 5200
Title: Inferential Replicability for Online Decision-Making Algorithms
Abstract: Reinforcement learning (RL) algorithms are increasingly used as an intervention component in digital health interventions. Including the RL algorithm in an intervention can complicate data analyses on the resulting longitudinal data. Longitudinal data analyses, and causal inference, more broadly are critical for deciding how to refine the intervention, and deciding whether to further deploy the intervention. This talk discusses the replicability of statistical analyses in these settings. We demonstrate that many standard statistical estimators can be inconsistent and fail to be replicable across repetitions of the deployment, even as the sample size grows large. We show that this non-replicability is intimately related to properties of the RL algorithm itself. We introduce a formal definition of a “replicable RL algorithm” and prove that under such algorithms, a wide variety of common statistical analyses are guaranteed to be consistent. This work is motivated by our involvement in a mobile health oral health self-care intervention and a mobile intervention designed to help people reduce cannabis use.
Bio: Susan Murphy’s research focuses on improving sequential, individualized, decision making in digital health. She developed the micro-randomized trial for use in constructing digital health interventions; this trial design is in use across a broad range of health-related areas. Her lab develops and deploys online learning algorithms for developing personalized digital health interventions. Dr. Murphy is a member of the National Academy of Sciences and of the National Academy of Medicine, both of the US National Academies. In 2013 she was awarded a MacArthur Fellowship for her work on experimental designs to inform sequential decision making. She is a Fellow of the College on Problems in Drug Dependence, Past-President of Institute of Mathematical Statistics, Past-President of the Bernoulli Society and a former editor of the Annals of Statistics.
See this link for more details of the seminar: https://sites.google.com/g.ucla.edu/data-theory-seminar