Classification uncertainty for transient gravitational-wave noise artifacts with optimized conformal prediction
Classification uncertainty for transient gravitational-wave noise artifacts with optimized conformal prediction
With the increasing use of machine learning (ML) algorithms in scientific research comes the need for reliable uncertainty quantification. When taking a measurement it is not enough to provide the result, we also have to declare how confident we are in the measurement. This is also true when the results are obtained from a ML algorithm, and arguably more so since the internal workings of ML algorithms are often less transparent compared to traditional statistical methods. Additionally, many ML algorithms do not provide uncertainty estimates, and auxiliary algorithms must be applied. Conformal prediction (
Malz, Ann-Kristin
4d1e6a33-2361-4fc6-8329-814d76027a05
Ashton, Gregory
a8cec4b1-3c98-4b28-af2a-1e37cb3b9f2a
Colombo, Nicolo
954ff6c1-95e2-493b-b64a-0b6b222a70dd
29 April 2025
Malz, Ann-Kristin
4d1e6a33-2361-4fc6-8329-814d76027a05
Ashton, Gregory
a8cec4b1-3c98-4b28-af2a-1e37cb3b9f2a
Colombo, Nicolo
954ff6c1-95e2-493b-b64a-0b6b222a70dd
Malz, Ann-Kristin, Ashton, Gregory and Colombo, Nicolo
(2025)
Classification uncertainty for transient gravitational-wave noise artifacts with optimized conformal prediction.
Physical Review D, 111 (8), [084078].
(doi:10.1103/PhysRevD.111.084078).
Abstract
With the increasing use of machine learning (ML) algorithms in scientific research comes the need for reliable uncertainty quantification. When taking a measurement it is not enough to provide the result, we also have to declare how confident we are in the measurement. This is also true when the results are obtained from a ML algorithm, and arguably more so since the internal workings of ML algorithms are often less transparent compared to traditional statistical methods. Additionally, many ML algorithms do not provide uncertainty estimates, and auxiliary algorithms must be applied. Conformal prediction (
Text
PhysRevD.111.084078
- Version of Record
More information
Accepted/In Press date: 7 April 2025
Published date: 29 April 2025
Identifiers
Local EPrints ID: 508290
URI: http://eprints.soton.ac.uk/id/eprint/508290
ISSN: 1550-7998
PURE UUID: 3db590e3-4052-4dab-a4c4-28001ebb2fa0
Catalogue record
Date deposited: 16 Jan 2026 17:32
Last modified: 17 Jan 2026 03:47
Export record
Altmetrics
Contributors
Author:
Ann-Kristin Malz
Author:
Gregory Ashton
Author:
Nicolo Colombo
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics