Learning causal structures using regression invariance

Amir Emad Ghassami, Saber Salehkaleybar, Negar Kiyavash, Kun Zhang

Research output: Contribution to journalConference article

Abstract

We study causal discovery in a multi-environment setting, in which the functional relations for producing the variables from their direct causes remain the same across environments, while the distribution of exogenous noises may vary. We introduce the idea of using the invariance of the functional relations of the variables to their causes across a set of environments for structure learning. We define a notion of completeness for a causal inference algorithm in this setting and prove the existence of such algorithm by proposing the baseline algorithm. Additionally, we present an alternate algorithm that has significantly improved computational and sample complexity compared to the baseline algorithm. Experiment results show that the proposed algorithm outperforms the other existing algorithms.

Original languageEnglish (US)
Pages (from-to)3012-3022
Number of pages11
JournalAdvances in Neural Information Processing Systems
Volume2017-December
StatePublished - Jan 1 2017
Event31st Annual Conference on Neural Information Processing Systems, NIPS 2017 - Long Beach, United States
Duration: Dec 4 2017Dec 9 2017

Fingerprint

Invariance
Experiments

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Ghassami, A. E., Salehkaleybar, S., Kiyavash, N., & Zhang, K. (2017). Learning causal structures using regression invariance. Advances in Neural Information Processing Systems, 2017-December, 3012-3022.

Learning causal structures using regression invariance. / Ghassami, Amir Emad; Salehkaleybar, Saber; Kiyavash, Negar; Zhang, Kun.

In: Advances in Neural Information Processing Systems, Vol. 2017-December, 01.01.2017, p. 3012-3022.

Research output: Contribution to journalConference article

Ghassami, AE, Salehkaleybar, S, Kiyavash, N & Zhang, K 2017, 'Learning causal structures using regression invariance', Advances in Neural Information Processing Systems, vol. 2017-December, pp. 3012-3022.
Ghassami AE, Salehkaleybar S, Kiyavash N, Zhang K. Learning causal structures using regression invariance. Advances in Neural Information Processing Systems. 2017 Jan 1;2017-December:3012-3022.
Ghassami, Amir Emad ; Salehkaleybar, Saber ; Kiyavash, Negar ; Zhang, Kun. / Learning causal structures using regression invariance. In: Advances in Neural Information Processing Systems. 2017 ; Vol. 2017-December. pp. 3012-3022.
@article{32947d2710fa4d9289a902431dfea713,
title = "Learning causal structures using regression invariance",
abstract = "We study causal discovery in a multi-environment setting, in which the functional relations for producing the variables from their direct causes remain the same across environments, while the distribution of exogenous noises may vary. We introduce the idea of using the invariance of the functional relations of the variables to their causes across a set of environments for structure learning. We define a notion of completeness for a causal inference algorithm in this setting and prove the existence of such algorithm by proposing the baseline algorithm. Additionally, we present an alternate algorithm that has significantly improved computational and sample complexity compared to the baseline algorithm. Experiment results show that the proposed algorithm outperforms the other existing algorithms.",
author = "Ghassami, {Amir Emad} and Saber Salehkaleybar and Negar Kiyavash and Kun Zhang",
year = "2017",
month = "1",
day = "1",
language = "English (US)",
volume = "2017-December",
pages = "3012--3022",
journal = "Advances in Neural Information Processing Systems",
issn = "1049-5258",

}

TY - JOUR

T1 - Learning causal structures using regression invariance

AU - Ghassami, Amir Emad

AU - Salehkaleybar, Saber

AU - Kiyavash, Negar

AU - Zhang, Kun

PY - 2017/1/1

Y1 - 2017/1/1

N2 - We study causal discovery in a multi-environment setting, in which the functional relations for producing the variables from their direct causes remain the same across environments, while the distribution of exogenous noises may vary. We introduce the idea of using the invariance of the functional relations of the variables to their causes across a set of environments for structure learning. We define a notion of completeness for a causal inference algorithm in this setting and prove the existence of such algorithm by proposing the baseline algorithm. Additionally, we present an alternate algorithm that has significantly improved computational and sample complexity compared to the baseline algorithm. Experiment results show that the proposed algorithm outperforms the other existing algorithms.

AB - We study causal discovery in a multi-environment setting, in which the functional relations for producing the variables from their direct causes remain the same across environments, while the distribution of exogenous noises may vary. We introduce the idea of using the invariance of the functional relations of the variables to their causes across a set of environments for structure learning. We define a notion of completeness for a causal inference algorithm in this setting and prove the existence of such algorithm by proposing the baseline algorithm. Additionally, we present an alternate algorithm that has significantly improved computational and sample complexity compared to the baseline algorithm. Experiment results show that the proposed algorithm outperforms the other existing algorithms.

UR - http://www.scopus.com/inward/record.url?scp=85047011824&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047011824&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85047011824

VL - 2017-December

SP - 3012

EP - 3022

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

SN - 1049-5258

ER -