[2603.02289] Topological Causal Effects

Statistics > Methodology arXiv:2603.02289 (stat) [Submitted on 2 Mar 2026] Title:Topological Causal Effects Authors:Kwangho Kim, Hajin Lee View a PDF of the paper titled Topological Causal Effects, by Kwangho Kim and Hajin Lee View PDF HTML (experimental) Abstract:Estimating causal effects is particularly challenging when outcomes arise in complex, non-Euclidean spaces, where conventional methods often fail to capture meaningful structural variation. We develop a framework for topological causal inference that defines treatment effects through differences in the topological structure of potential outcomes, summarized by power-weighted silhouette functions of persistence diagrams. We develop an efficient, doubly robust estimator in a fully nonparametric model, establish functional weak convergence, and construct a formal test of the null hypothesis of no topological effect. Empirical studies illustrate that the proposed method reliably quantifies topological treatment effects across diverse complex outcome types. Subjects: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML) Cite as: arXiv:2603.02289 [stat.ME]   (or arXiv:2603.02289v1 [stat.ME] for this version)   https://doi.org/10.48550/arXiv.2603.02289 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journal reference: Proceedings of the Fourteenth International Conference on Learning Representations (ICLR 2026) Submission history From: Kwangho Kim [view email] [v1] Mon, 2 M...