Skip to yearly menu bar Skip to main content


Poster

DiffPO: A causal diffusion model for learning distributions of potential outcomes

Yuchen Ma · Valentyn Melnychuk · Jonas Schweisthal · Stefan Feuerriegel

West Ballroom A-D #6507
[ ]
Wed 11 Dec 11 a.m. PST — 2 p.m. PST

Abstract:

Predicting potential outcomes of interventions from observational data is crucial for decision-making in medicine, but the task is challenging due to the fundamental problem of causal inference. Existing methods are largely limited to point estimates of potential outcomes with no uncertain quantification; thus, the full information about the distributions of potential outcomes is typically ignored. In this paper, we propose a novel causal diffusion model called DiffPO, which is carefully designed for reliable inferences in medicine by learning the distribution of potential outcomes. In our DiffPO, we leverage a tailored conditional denoising diffusion model to learn complex distributions, where we address the selection bias through a novel orthogonal diffusion loss. Another strength of our DiffPO method is that it is highly flexible (e.g., it can also be used to estimate different causal quantities such as CATE). Across a wide range of experiments, we show that our method achieves state-of-the-art performance.

Live content is unavailable. Log in and register to view live content