Identifiability of Sparse Causal Effects using Instrumental Variables

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt

Standard

Identifiability of Sparse Causal Effects using Instrumental Variables. / Pfister, Niklas; Peters, Jonas.

Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. PMLR, 2022. s. 1613-1622 (Proceedings of Machine Learning Research, Bind 180).

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt

Harvard

Pfister, N & Peters, J 2022, Identifiability of Sparse Causal Effects using Instrumental Variables. i Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. PMLR, Proceedings of Machine Learning Research, bind 180, s. 1613-1622, 38th Conference on Uncertainty in Artificial Intelligence, UAI 2022, Eindhoven, Holland, 01/08/2022. <https://proceedings.mlr.press/v180/pfister22a.html>

APA

Pfister, N., & Peters, J. (2022). Identifiability of Sparse Causal Effects using Instrumental Variables. I Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence (s. 1613-1622). PMLR. Proceedings of Machine Learning Research Bind 180 https://proceedings.mlr.press/v180/pfister22a.html

Vancouver

Pfister N, Peters J. Identifiability of Sparse Causal Effects using Instrumental Variables. I Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. PMLR. 2022. s. 1613-1622. (Proceedings of Machine Learning Research, Bind 180).

Author

Pfister, Niklas ; Peters, Jonas. / Identifiability of Sparse Causal Effects using Instrumental Variables. Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence. PMLR, 2022. s. 1613-1622 (Proceedings of Machine Learning Research, Bind 180).

Bibtex

@inproceedings{9e2176b168704ec9936388d162e9f53b,

title = "Identifiability of Sparse Causal Effects using Instrumental Variables",

abstract = "Exogenous heterogeneity, for example, in the form of instrumental variables can help us learn a system's underlying causal structure and predict the outcome of unseen intervention experiments. In this paper, we consider linear models in which the causal effect from covariates X on a response Y is sparse. We provide conditions under which the causal coefficient becomes identifiable from the observed distribution. These conditions can be satisfied even if the number of instruments is as small as the number of causal parents. We also develop graphical criteria under which identifiability holds with probability one if the edge coefficients are sampled randomly from a distribution that is absolutely continuous with respect to Lebesgue measure and Y is childless. As an estimator, we propose spaceIV and prove that it consistently estimates the causal effect if the model is identifiable and evaluate its performance on simulated data. If identifiability does not hold, we show that it may still be possible to recover a subset of the causal parents.",

author = "Niklas Pfister and Jonas Peters",

year = "2022",

language = "English",

series = "Proceedings of Machine Learning Research",

pages = "1613--1622",

booktitle = "Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence",

publisher = "PMLR",

}

RIS

TY - GEN

T1 - Identifiability of Sparse Causal Effects using Instrumental Variables

AU - Pfister, Niklas

AU - Peters, Jonas

PY - 2022

Y1 - 2022

N2 - Exogenous heterogeneity, for example, in the form of instrumental variables can help us learn a system's underlying causal structure and predict the outcome of unseen intervention experiments. In this paper, we consider linear models in which the causal effect from covariates X on a response Y is sparse. We provide conditions under which the causal coefficient becomes identifiable from the observed distribution. These conditions can be satisfied even if the number of instruments is as small as the number of causal parents. We also develop graphical criteria under which identifiability holds with probability one if the edge coefficients are sampled randomly from a distribution that is absolutely continuous with respect to Lebesgue measure and Y is childless. As an estimator, we propose spaceIV and prove that it consistently estimates the causal effect if the model is identifiable and evaluate its performance on simulated data. If identifiability does not hold, we show that it may still be possible to recover a subset of the causal parents.

AB - Exogenous heterogeneity, for example, in the form of instrumental variables can help us learn a system's underlying causal structure and predict the outcome of unseen intervention experiments. In this paper, we consider linear models in which the causal effect from covariates X on a response Y is sparse. We provide conditions under which the causal coefficient becomes identifiable from the observed distribution. These conditions can be satisfied even if the number of instruments is as small as the number of causal parents. We also develop graphical criteria under which identifiability holds with probability one if the edge coefficients are sampled randomly from a distribution that is absolutely continuous with respect to Lebesgue measure and Y is childless. As an estimator, we propose spaceIV and prove that it consistently estimates the causal effect if the model is identifiable and evaluate its performance on simulated data. If identifiability does not hold, we show that it may still be possible to recover a subset of the causal parents.

UR - http://www.scopus.com/inward/record.url?scp=85163350451&partnerID=8YFLogxK

M3 - Article in proceedings

AN - SCOPUS:85163350451

T3 - Proceedings of Machine Learning Research

SP - 1613

EP - 1622

BT - Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence

PB - PMLR

T2 - 38th Conference on Uncertainty in Artificial Intelligence, UAI 2022

Y2 - 1 August 2022 through 5 August 2022

ER -

ID: 359652751

Institut for Matematiske Fag