Learning Joint Interventional Effects from Single-Variable Interventions in Additive Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kekić, Armin, Mejia, Sergio Hernan Garrido, Schölkopf, Bernhard
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910989448904704
author Kekić, Armin
Mejia, Sergio Hernan Garrido
Schölkopf, Bernhard
author_facet Kekić, Armin
Mejia, Sergio Hernan Garrido
Schölkopf, Bernhard
contents Estimating causal effects of joint interventions on multiple variables is crucial in many domains, but obtaining data from such simultaneous interventions can be challenging. Our study explores how to learn joint interventional effects using only observational data and single-variable interventions. We present an identifiability result for this problem, showing that for a class of nonlinear additive outcome mechanisms, joint effects can be inferred without access to joint interventional data. We propose a practical estimator that decomposes the causal effect into confounded and unconfounded contributions for each intervention variable. Experiments on synthetic data demonstrate that our method achieves performance comparable to models trained directly on joint interventional data, outperforming a purely observational estimator.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04945
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Joint Interventional Effects from Single-Variable Interventions in Additive Models
Kekić, Armin
Mejia, Sergio Hernan Garrido
Schölkopf, Bernhard
Machine Learning
Estimating causal effects of joint interventions on multiple variables is crucial in many domains, but obtaining data from such simultaneous interventions can be challenging. Our study explores how to learn joint interventional effects using only observational data and single-variable interventions. We present an identifiability result for this problem, showing that for a class of nonlinear additive outcome mechanisms, joint effects can be inferred without access to joint interventional data. We propose a practical estimator that decomposes the causal effect into confounded and unconfounded contributions for each intervention variable. Experiments on synthetic data demonstrate that our method achieves performance comparable to models trained directly on joint interventional data, outperforming a purely observational estimator.
title Learning Joint Interventional Effects from Single-Variable Interventions in Additive Models
topic Machine Learning
url https://arxiv.org/abs/2506.04945