Saved in:
Bibliographic Details
Main Authors: Cai, Ming, Gao, Penggang, Hara, Hisayuki
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.14780
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918161923702784
author Cai, Ming
Gao, Penggang
Hara, Hisayuki
author_facet Cai, Ming
Gao, Penggang
Hara, Hisayuki
contents This paper addresses the problem of estimating causal directed acyclic graphs in linear non-Gaussian acyclic models with latent confounders (LvLiNGAM). Existing methods assume mutually independent latent confounders or cannot properly handle models with causal relationships among observed variables. We propose a novel algorithm that identifies causal DAGs in LvLiNGAM, allowing causal structures among latent variables, among observed variables, and between the two. The proposed method leverages higher-order cumulants of observed data to identify the causal structure. Extensive simulations and experiments with real-world data demonstrate the validity and practical utility of the proposed algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2510_14780
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Causal Discovery for Linear DAGs with Dependent Latent Variables via Higher-order Cumulants
Cai, Ming
Gao, Penggang
Hara, Hisayuki
Machine Learning
This paper addresses the problem of estimating causal directed acyclic graphs in linear non-Gaussian acyclic models with latent confounders (LvLiNGAM). Existing methods assume mutually independent latent confounders or cannot properly handle models with causal relationships among observed variables. We propose a novel algorithm that identifies causal DAGs in LvLiNGAM, allowing causal structures among latent variables, among observed variables, and between the two. The proposed method leverages higher-order cumulants of observed data to identify the causal structure. Extensive simulations and experiments with real-world data demonstrate the validity and practical utility of the proposed algorithm.
title Causal Discovery for Linear DAGs with Dependent Latent Variables via Higher-order Cumulants
topic Machine Learning
url https://arxiv.org/abs/2510.14780