[2603.02159] Instrumental and Proximal Causal Inference with Gaussian Processes

Statistics > Machine Learning arXiv:2603.02159 (stat) [Submitted on 2 Mar 2026] Title:Instrumental and Proximal Causal Inference with Gaussian Processes Authors:Yuqi Zhang, Krikamol Muandet, Dino Sejdinovic, Edwin Fong, Siu Lun Chau View a PDF of the paper titled Instrumental and Proximal Causal Inference with Gaussian Processes, by Yuqi Zhang and 4 other authors View PDF Abstract:Instrumental variable (IV) and proximal causal learning (Proxy) methods are central frameworks for causal inference in the presence of unobserved confounding. Despite substantial methodological advances, existing approaches rarely provide reliable epistemic uncertainty (EU) quantification. We address this gap through a Deconditional Gaussian Process (DGP) framework for uncertainty-aware causal learning. Our formulation recovers popular kernel estimators as the posterior mean, ensuring predictive precision, while the posterior variance yields principled and well-calibrated EU. Moreover, the probabilistic structure enables systematic model selection via marginal log-likelihood optimization. Empirical results demonstrate strong predictive performance alongside informative EU quantification, evaluated via empirical coverage frequencies and decision-aware accuracy rejection curves. Together, our approach provides a unified, practical solution for causal inference under unobserved confounding with reliable uncertainty. Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG) Cite as: arXiv:2603.0...