Causality-Encoded Diffusion Models for Interventional Sampling and Edge Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Li, Shen, Xiaotong, Pan, Wei
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917431379755008
author Chen, Li
Shen, Xiaotong
Pan, Wei
author_facet Chen, Li
Shen, Xiaotong
Pan, Wei
contents Standard diffusion models are flexible estimators of complex distributions, but they do not encode causal structures and therefore do not by themselves support causal analysis. We propose a causality-encoded diffusion framework that incorporates a known directed acyclic graph by training conditional diffusion models consistent with the graph factorisation. The resulting sampler approximately recovers the observational distribution and enables interventional sampling by fixing intervened variables while propagating effects through the graph during reverse diffusion. Building on this interventional simulator, we develop a resampling-based test for directed edges that generates null replicates under a candidate graph. We establish convergence guarantees for observational and interventional distribution estimation, with rates governed by the maximum local dimension rather than the ambient dimension, and prove asymptotic control of type I error for the edge test. Simulations show improved interventional distribution recovery relative to baselines, with near-nominal size and favourable power in inference. An application to flow cytometry data demonstrates practical utility of the proposed method in assessing disputed signalling linkages.
format Preprint
id arxiv_https___arxiv_org_abs_2604_21843
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Causality-Encoded Diffusion Models for Interventional Sampling and Edge Inference
Chen, Li
Shen, Xiaotong
Pan, Wei
Methodology
Machine Learning
Standard diffusion models are flexible estimators of complex distributions, but they do not encode causal structures and therefore do not by themselves support causal analysis. We propose a causality-encoded diffusion framework that incorporates a known directed acyclic graph by training conditional diffusion models consistent with the graph factorisation. The resulting sampler approximately recovers the observational distribution and enables interventional sampling by fixing intervened variables while propagating effects through the graph during reverse diffusion. Building on this interventional simulator, we develop a resampling-based test for directed edges that generates null replicates under a candidate graph. We establish convergence guarantees for observational and interventional distribution estimation, with rates governed by the maximum local dimension rather than the ambient dimension, and prove asymptotic control of type I error for the edge test. Simulations show improved interventional distribution recovery relative to baselines, with near-nominal size and favourable power in inference. An application to flow cytometry data demonstrates practical utility of the proposed method in assessing disputed signalling linkages.
title Causality-Encoded Diffusion Models for Interventional Sampling and Edge Inference
topic Methodology
Machine Learning
url https://arxiv.org/abs/2604.21843