Astronomaly at Scale: Searching for Anomalies Amongst 4 Million Galaxies

Find out the latest thinking about our universe.
User avatar
Ann
4725 Å
Posts: 13551
Joined: Sat May 29, 2010 5:33 am

Astronomaly at Scale: Searching for Anomalies Amongst 4 Million Galaxies

Post by Ann » Tue Oct 03, 2023 4:44 am

Anomalous galaxies Astronomaly.png
V. Etsebeth et al. wrote:
Modern astronomical surveys are producing datasets of unprecedented size and richness, increasing the potential for high-impact scientific discovery. This possibility, coupled with the challenge of exploring a large number of sources, has led to the development of novel machine-learning-based anomaly detection approaches, such as astronomaly. For the first time, we test the scalability of astronomaly by applying it to almost 4 million images of galaxies from the Dark Energy Camera Legacy Survey. We use a trained deep learning algorithm to learn useful representations of the images and pass these to the anomaly detection algorithm isolation forest, coupled with astronomaly’s active learning method, to discover interesting sources. We find that data selection criteria have a significant impact on the trade-off between finding rare sources such as strong lenses and introducing artefacts into the dataset. We demonstrate that active learning is required to identify the most interesting sources and reduce artefacts, while anomaly detection methods alone are insufficient. Using astronomaly, we find 1635 anomalies among the top 2000 sources in the dataset after applying active learning, including 8 strong gravitational lens candidates, 1609 galaxy merger candidates, and 18 previously unidentified sources exhibiting highly unusual morphology. Our results show that by leveraging the human-machine interface, astronomaly is able to rapidly identify sources of scientific interest even in large datasets.
Astronamly is about using machine learning to find interesting galaxies. At right you can see 18 "anomalous" galaxies found by Astronomaly. Personally I'm fascinated by U1, the "6 in the sky". I like U3, too, which looks like an edge-on galaxy with an anomalous cyan center with vertical dust lanes, and a disk divided in half with radically different colors. But U17 mostly looks like the Siamese Twins Galaxies to me, although with a lot more star formation.

Strong lenses and fascinating mergers have also been found by Astronomaly.

Ann
You do not have the required permissions to view the files attached to this post.
Color Commentator

Waryin1956
Asternaut
Posts: 2
Joined: Fri Oct 13, 2023 9:36 am

Re: Astronomaly at Scale: Searching for Anomalies Amongst 4 Million Galaxies

Post by Waryin1956 » Fri Oct 13, 2023 9:39 am

I also like U1. Look like number 6 in the galaxy.

Waryin1956
Asternaut
Posts: 2
Joined: Fri Oct 13, 2023 9:36 am

Re: Astronomaly at Scale: Searching for Anomalies Amongst 4 Million Galaxies

Post by Waryin1956 » Mon Oct 16, 2023 1:06 am

The U6 looks like a face.fnaf games

saturno2
Commander
Posts: 759
Joined: Wed Jun 29, 2011 10:05 pm

Re: Astronomaly at Scale: Searching for Anomalies Amongst 4 Million Galaxies

Post by saturno2 » Sun Nov 05, 2023 2:03 am

Very interesting topic