A deep-learning search for technosignatures from 820 nearby stars

The goal of the search for extraterrestrial intelligence (SETI) is to quantify the prevalence of technological life beyond Earth via their ‘technosignatures’. One theorized technosignature is narrowband Doppler drifting radio signals. The principal challenge in conducting SETI in the radio domain is...

Full description

Saved in:
Bibliographic Details
Published in:Nature astronomy Vol. 7; no. 4; pp. 492 - 502
Main Authors: Ma, Peter Xiangyuan, Ng, Cherry, Rizk, Leandro, Croft, Steve, Siemion, Andrew P. V, Brzycki, Bryan, Czech, Daniel, Drew, Jamie, Gajjar, Vishal, Hoang, John, Isaacson, Howard, Lebofsky, Matt, MacMahon, David H. E, de Pater, Imke, Price, Danny C, Sheikh, Sofia Z, Worden, S. Pete
Format: Journal Article
Language:English
Published: London Nature Publishing Group 01.04.2023
Subjects:
Astronomy
Astrophysics
Candidates
Datasets
Deep learning
Initiatives
Intelligence
Machine learning
Neural networks
Stars
Telescopes
ISSN:2397-3366
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The goal of the search for extraterrestrial intelligence (SETI) is to quantify the prevalence of technological life beyond Earth via their ‘technosignatures’. One theorized technosignature is narrowband Doppler drifting radio signals. The principal challenge in conducting SETI in the radio domain is developing a generalized technique to reject human radiofrequency interference. Here we present a comprehensive deep-learning-based technosignature search on 820 stellar targets from the Hipparcos catalogue, totalling over 480 h of on-sky data taken with the Robert C. Byrd Green Bank Telescope as part of the Breakthrough Listen initiative. We implement a novel β-convolutional variational autoencoder to identify technosignature candidates in a semi-unsupervised manner while keeping the false-positive rate manageably low, reducing the number of candidate signals by approximately two orders of magnitude compared with previous analyses on the same dataset. Our work also returned eight promising extraterrestrial intelligence signals of interest not previously identified. Re-observations on these targets have so far not resulted in re-detections of signals with similar morphology. This machine-learning approach presents itself as a leading solution in accelerating SETI and other transient research into the age of data-driven astronomy.A state-of-the-art machine-learning method combs a 480-h-long dataset of 820 nearby stars from the SETI Breakthrough Listen project, reducing the number of interesting signals by two orders of magnitude. Further visual inspection identifies eight promising signals of interest from different stars that warrant further observations.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2397-3366
DOI:10.1038/s41550-022-01872-z