Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wu, Rui, Xie, Hong, Li, Yongjun
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2603.17385
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911525107662848
author Wu, Rui
Xie, Hong
Li, Yongjun
author_facet Wu, Rui
Xie, Hong
Li, Yongjun
contents Judea Pearl's do-calculus provides a foundation for causal inference, but its translation to continuous generative models remains fraught with geometric challenges. We establish the fundamental limits of such interventions. We define the Counterfactual Event Horizon and prove the Manifold Tearing Theorem: deterministic flows inevitably develop finite-time singularities under extreme interventions. We establish the Causal Uncertainty Principle for the trade-off between intervention extremity and identity preservation. Finally, we introduce Geometry-Aware Causal Flow (GACF), a scalable algorithm that utilizes a topological radar to bypass manifold tearing, validated on high-dimensional scRNA-seq data.
format Preprint
id arxiv_https___arxiv_org_abs_2603_17385
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Causal Uncertainty Principle: Manifold Tearing and the Topological Limits of Counterfactual Interventions
Wu, Rui
Xie, Hong
Li, Yongjun
Machine Learning
62A01, 49Q22, 60H10
I.2.0; G.3
Judea Pearl's do-calculus provides a foundation for causal inference, but its translation to continuous generative models remains fraught with geometric challenges. We establish the fundamental limits of such interventions. We define the Counterfactual Event Horizon and prove the Manifold Tearing Theorem: deterministic flows inevitably develop finite-time singularities under extreme interventions. We establish the Causal Uncertainty Principle for the trade-off between intervention extremity and identity preservation. Finally, we introduce Geometry-Aware Causal Flow (GACF), a scalable algorithm that utilizes a topological radar to bypass manifold tearing, validated on high-dimensional scRNA-seq data.
title The Causal Uncertainty Principle: Manifold Tearing and the Topological Limits of Counterfactual Interventions
topic Machine Learning
62A01, 49Q22, 60H10
I.2.0; G.3
url https://arxiv.org/abs/2603.17385