Generative Intervention Models for Causal Perturbation Modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schneider, Nora, Lorch, Lars, Kilbertus, Niki, Schölkopf, Bernhard, Krause, Andreas
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916817640882176
author Schneider, Nora
Lorch, Lars
Kilbertus, Niki
Schölkopf, Bernhard
Krause, Andreas
author_facet Schneider, Nora
Lorch, Lars
Kilbertus, Niki
Schölkopf, Bernhard
Krause, Andreas
contents We consider the problem of predicting perturbation effects via causal models. In many applications, it is a priori unknown which mechanisms of a system are modified by an external perturbation, even though the features of the perturbation are available. For example, in genomics, some properties of a drug may be known, but not their causal effects on the regulatory pathways of cells. We propose a generative intervention model (GIM) that learns to map these perturbation features to distributions over atomic interventions in a jointly-estimated causal model. Contrary to prior approaches, this enables us to predict the distribution shifts of unseen perturbation features while gaining insights about their mechanistic effects in the underlying data-generating process. On synthetic data and scRNA-seq drug perturbation data, GIMs achieve robust out-of-distribution predictions on par with unstructured approaches, while effectively inferring the underlying perturbation mechanisms, often better than other causal inference methods.
format Preprint
id arxiv_https___arxiv_org_abs_2411_14003
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Generative Intervention Models for Causal Perturbation Modeling
Schneider, Nora
Lorch, Lars
Kilbertus, Niki
Schölkopf, Bernhard
Krause, Andreas
Machine Learning
We consider the problem of predicting perturbation effects via causal models. In many applications, it is a priori unknown which mechanisms of a system are modified by an external perturbation, even though the features of the perturbation are available. For example, in genomics, some properties of a drug may be known, but not their causal effects on the regulatory pathways of cells. We propose a generative intervention model (GIM) that learns to map these perturbation features to distributions over atomic interventions in a jointly-estimated causal model. Contrary to prior approaches, this enables us to predict the distribution shifts of unseen perturbation features while gaining insights about their mechanistic effects in the underlying data-generating process. On synthetic data and scRNA-seq drug perturbation data, GIMs achieve robust out-of-distribution predictions on par with unstructured approaches, while effectively inferring the underlying perturbation mechanisms, often better than other causal inference methods.
title Generative Intervention Models for Causal Perturbation Modeling
topic Machine Learning
url https://arxiv.org/abs/2411.14003