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

TL;DR

This paper introduces a method to estimate joint causal effects from observational data and single-variable interventions in nonlinear additive models, enabling inference without joint intervention data.

Contribution

It provides an identifiability result and a practical estimator for joint effects using only single-variable intervention data in nonlinear additive models.

Findings

Estimator achieves performance comparable to models trained on joint data

Outperforms purely observational estimators

Validates approach on synthetic data

Abstract

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.