CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series

The study of cause and effect is of the utmost importance in many branches of science, but also for many practical applications of intelligent systems. In particular, identifying causal relationships in situations that include hidden factors is a major challenge for methods that rely solely on obser...

Full description

Saved in:
Bibliographic Details
Main Authors: Luca Castri, Sariah Mghames, Marc Hanheide, Nicola Bellotto
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:Advanced Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1002/aisy.202400181
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850243578481082368
author Luca Castri
Sariah Mghames
Marc Hanheide
Nicola Bellotto
author_facet Luca Castri
Sariah Mghames
Marc Hanheide
Nicola Bellotto
author_sort Luca Castri
collection DOAJ
description The study of cause and effect is of the utmost importance in many branches of science, but also for many practical applications of intelligent systems. In particular, identifying causal relationships in situations that include hidden factors is a major challenge for methods that rely solely on observational data for building causal models. This article proposes CAnDOIT, a causal discovery method to reconstruct causal models using both observational and interventional time‐series data. The use of interventional data in the causal analysis is crucial for real‐world applications, such as robotics, where the scenario is highly complex and observational data alone are often insufficient to uncover the correct causal structure. Validation of the method is performed initially on randomly generated synthetic models and subsequently on a well‐known benchmark for causal structure learning in a robotic manipulation environment. The experiments demonstrate that the approach can effectively handle data from interventions and exploit them to enhance the accuracy of the causal analysis. A Python implementation of CAnDOIT is developed and is publicly available on GitHub: https://github.com/lcastri/causalflow.
format Article
id doaj-art-059f520a12c34db5b05dc60a0c04fc76
institution OA Journals
issn 2640-4567
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series Advanced Intelligent Systems
spelling doaj-art-059f520a12c34db5b05dc60a0c04fc762025-08-20T01:59:57ZengWileyAdvanced Intelligent Systems2640-45672024-12-01612n/an/a10.1002/aisy.202400181CAnDOIT: Causal Discovery with Observational and Interventional Data from Time SeriesLuca Castri0Sariah Mghames1Marc Hanheide2Nicola Bellotto3L‐CAS, School of Engineering & Physical Sciences University of Lincoln Lincoln LN6 7TS UKL‐CAS, School of Engineering & Physical Sciences University of Lincoln Lincoln LN6 7TS UKL‐CAS, School of Engineering & Physical Sciences University of Lincoln Lincoln LN6 7TS UKL‐CAS, School of Engineering & Physical Sciences University of Lincoln Lincoln LN6 7TS UKThe study of cause and effect is of the utmost importance in many branches of science, but also for many practical applications of intelligent systems. In particular, identifying causal relationships in situations that include hidden factors is a major challenge for methods that rely solely on observational data for building causal models. This article proposes CAnDOIT, a causal discovery method to reconstruct causal models using both observational and interventional time‐series data. The use of interventional data in the causal analysis is crucial for real‐world applications, such as robotics, where the scenario is highly complex and observational data alone are often insufficient to uncover the correct causal structure. Validation of the method is performed initially on randomly generated synthetic models and subsequently on a well‐known benchmark for causal structure learning in a robotic manipulation environment. The experiments demonstrate that the approach can effectively handle data from interventions and exploit them to enhance the accuracy of the causal analysis. A Python implementation of CAnDOIT is developed and is publicly available on GitHub: https://github.com/lcastri/causalflow.https://doi.org/10.1002/aisy.202400181causal roboticsobservations and interventions‐based causal discoveriestime series
spellingShingle Luca Castri
Sariah Mghames
Marc Hanheide
Nicola Bellotto
CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series
Advanced Intelligent Systems
causal robotics
observations and interventions‐based causal discoveries
time series
title CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series
title_full CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series
title_fullStr CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series
title_full_unstemmed CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series
title_short CAnDOIT: Causal Discovery with Observational and Interventional Data from Time Series
title_sort candoit causal discovery with observational and interventional data from time series
topic causal robotics
observations and interventions‐based causal discoveries
time series
url https://doi.org/10.1002/aisy.202400181
work_keys_str_mv AT lucacastri candoitcausaldiscoverywithobservationalandinterventionaldatafromtimeseries
AT sariahmghames candoitcausaldiscoverywithobservationalandinterventionaldatafromtimeseries
AT marchanheide candoitcausaldiscoverywithobservationalandinterventionaldatafromtimeseries
AT nicolabellotto candoitcausaldiscoverywithobservationalandinterventionaldatafromtimeseries