Saved in:
Bibliographic Details
Main Authors: Godon, Thibaud, Bauvin, Baptiste, Germain, Pascal, Corbeil, Jacques, Drouin, Alexandre
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.04777
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912286513299456
author Godon, Thibaud
Bauvin, Baptiste
Germain, Pascal
Corbeil, Jacques
Drouin, Alexandre
author_facet Godon, Thibaud
Bauvin, Baptiste
Germain, Pascal
Corbeil, Jacques
Drouin, Alexandre
contents Rule-based models, such as decision trees, appeal to practitioners due to their interpretable nature. However, the learning algorithms that produce such models are often vulnerable to spurious associations and thus, they are not guaranteed to extract causally-relevant insights. In this work, we build on ideas from the invariant causal prediction literature to propose Invariant Causal Set Covering Machines, an extension of the classical Set Covering Machine algorithm for conjunctions/disjunctions of binary-valued rules that provably avoids spurious associations. We demonstrate both theoretically and empirically that our method can identify the causal parents of a variable of interest in polynomial time.
format Preprint
id arxiv_https___arxiv_org_abs_2306_04777
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Invariant Causal Set Covering Machines
Godon, Thibaud
Bauvin, Baptiste
Germain, Pascal
Corbeil, Jacques
Drouin, Alexandre
Machine Learning
Methodology
Rule-based models, such as decision trees, appeal to practitioners due to their interpretable nature. However, the learning algorithms that produce such models are often vulnerable to spurious associations and thus, they are not guaranteed to extract causally-relevant insights. In this work, we build on ideas from the invariant causal prediction literature to propose Invariant Causal Set Covering Machines, an extension of the classical Set Covering Machine algorithm for conjunctions/disjunctions of binary-valued rules that provably avoids spurious associations. We demonstrate both theoretically and empirically that our method can identify the causal parents of a variable of interest in polynomial time.
title Invariant Causal Set Covering Machines
topic Machine Learning
Methodology
url https://arxiv.org/abs/2306.04777