Alex Luedtke - DoubleGen: Debiased Generative Modeling of Counterfactuals

Speaker: Alex Luedtke

 

Institution: Harvard Medical School

 

Title: DoubleGen: Debiased Generative Modeling of Counterfactuals

 

Abstract: 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.

 

Bio: Alex Luedtke is a statistician in Harvard Medical School's Department of Health Care Policy. From 2016 to 2025, he was a faculty member in the University of Washington's Department of Statistics and Fred Hutch's Vaccine and Infectious Disease Division. Alex received his ScB in Applied Mathematics from Brown University and his PhD in Biostatistics from the University of California, Berkeley.