ATTENZIONE! IL REPOSITORY E’ APERTO ESCLUSIVAMENTE ALL’IMMISSIONE DEI RAPPORTI TECNICI (COLLEZIONE 4.01). NON E’ POSSIBILE INSERIRE PRODOTTI IN TUTTE LE ALTRE COLLEZIONI, A CAUSA DI PROBLEMI TECNICI. THE REPOSITORY IS OPEN EXCLUSIVELY FOR THE SUBMISSION OF TECHNICAL REPORTS (COLLECTION 4.01). SUBMISSIONS TO ALL OTHER COLLECTIONS ARE TEMPORARILY UNAVAILABLE, DUE TO TECHNICAL ISSUES.
 

DeepGraviLens: a multi-modal architecture for classifying gravitational lensing data

Journal
Date Issued
2023
Author(s)
•
FRATERNALI, Piero
DOI
Abstract
Gravitational lensing is the relativistic effect generated by massive bodies, which bend the space-time surrounding them. It is a deeply investigated topic in astrophysics and allows validating theoretical relativistic results and studying faint astrophysical objects that would not be visible otherwise. In recent years, machine learning methods have been applied to support the analysis of the gravitational lensing phenomena by detecting lensing effects in datasets consisting of images associated with brightness variation time series. However, the state-of-the-art approaches either consider only images and neglect time-series data or achieve relatively low accuracy on the most difficult datasets. This paper introduces DeepGraviLens, a novel multi-modal network that classifies spatio-temporal data belonging to one non-lensed system type and three lensed system types. It surpasses the current state-of-the-art accuracy results by 3% to 11%, depending on the considered data set. Such an improvement will enable the acceleration of the analysis of lensed objects in upcoming astrophysical surveys, which will exploit the petabytes of data collected, e.g., from the Vera C. Rubin Observatory.
Uri
Url
Issn Identifier
0941-0643
Rights
open.access
File(s)
Loading...
Thumbnail Image
Name

s00521-023-08766-9 (1).pdf

Description
PDF editoriale
Size

1.57 MB

Format

Adobe PDF

Checksum (MD5)

d33663c6684aa3331797c0d92dd25854

Loading...
Thumbnail Image
Name

2205.00701.pdf

Description
preprint
Size

872.98 KB

Format

Adobe PDF

Checksum (MD5)

ec30cb3c57b675480c488c2389465d72