Hawes Hall, Classroom 101, Harvard Business School
DoubleGen: Debiased Generative Modeling of Counterfactuals
Generative models for counterfactual outcomes face two key sources of bias. Confounding bias arises when approaches fail to account for systematic differences between those who receive the intervention and those who do not. Misspecification bias arises when methods attempt to address confounding through estimation of an auxiliary model, but specify it incorrectly. We introduce DoubleGen, a doubly robust framework that modifies generative modeling training objectives to mitigate these biases. The new objectives rely on two auxiliaries — a propensity and outcome model — and successfully address confounding bias even if only one of them is correct. We provide finite-sample guarantees for this robustness property. We further establish conditions under which DoubleGen achieves oracle optimality — matching the convergence rates standard approaches would enjoy if interventional data were available — and minimax rate optimality. We illustrate DoubleGen with three examples: diffusion models, flow matching, and autoregressive language models.
Discussant: Rajarshi Mukherjee, Associate Professor of Biostatistics, Harvard T.H. Chan School of Public Health
Alex Luedtke
Statistician
Harvard Medical School
Alex Luedtke is a member of the faculty in the Department of Healthcare Policy at the Harvard Medical School and was an Associate Professor in the Department of Statistics at the University of Washington (UW). He also has an adjunct appointment in the Department of Biostatistics at UW and an affiliate appointment in the Vaccine and Infectious Disease Division at the Fred Hutchinson Cancer Research Center.
His methodological research focuses on developing efficient estimators in problems arising in a variety of areas, including in policy learning and infectious disease studies. He derives such estimators analytically using tools from semiparametric efficiency theory and numerically using minimax optimization schemes. He also serves as a study statistician for the HIV Vaccine Trials Network and the Covid-19 Prevention Network.
Alex is a recipient of an NIH Director’s New Innovator Award, AWS Machine Learning Research Award, Eric Lehmann Citation, and National Defense Science and Engineering Graduate (NDSEG) fellowship.