Work

Causal Discovery with Continuous Additive Noise Models

Year: 2013

Type: preprint

Abstract: We consider the problem of learning causal directed acyclic graphs from an observational joint distribution. One can use these graphs to predict the outcome of interventional experiments, from which data are often not available. We show that if the observational distribution follows a structural equation model with an additive noise structure, the directed acyclic graph becomes identifiable from the distribution under mild conditions. This constitutes an interesting alternative to traditional methods that assume faithfulness and identify only the Markov equivalence class of the graph, thus leaving some edges undirected. We provide practical algorithms for finitely many samples, RESIT (Regression with Subsequent Independence Test) and two methods based on an independence score. We prove that RESIT is correct in the population setting and provide an empirical evaluation.

Source: Journal of Machine Learning Research

DOI: 10.48550/arxiv.1309.6779

Authors: Peters, Jonas, Mooij, Joris, Janzing, Dominik, Schölkopf, Bernhard

Institutions: ETH Zurich, Seminar for Statistics, Radboud University Nijmegen, University of Amsterdam, Max Planck Institute for Intelligent Systems

Language: English

Topic: Bayesian Modeling and Causal Inference

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: European Commission, Nederlandse Organisatie voor Wetenschappelijk Onderzoek

Awards: G3207356259, G393177772, Causal Statistical Inference from High-Dimensional Data, G5702163051