Astronomers used artificial intelligence to find 1,400 “anomalous objects” from the Hubble archives

A pair of astronomers at the European Space Agency (ESA) have discovered more than 800 previously undocumented “astrophysical anomalies” hiding in the Hubble archives. To do this, researchers David O’Ryan and Pablo Gomez trained an AI model to comb through Hubble’s 35-year dataset, looking for UFOs and flagging them for manual review. “It is a treasure trove of data in which astrophysical anomalies may be found,” O’Ryan said in his article. statement.

Studying space is difficult. There are a lot of them, the place is noisy, and the deluge of data generated by instruments like the Hubble Space Telescope can overwhelm even large research teams. And sometimes space is weird. Very strange. Enter artificial intelligence, which is great at sifting through vast amounts of information to identify patterns, pointing out anomalies that astronomers might miss.

The model the astronomers used, called AnomalyMatch, scanned nearly 100 million cropped images from the Hubble Legacy Archive, marking the first time the dataset had been systematically searched for anomalies. Think of strangely shaped galaxies, light distorted by the gravity of massive objects, or planet-forming disks when viewed from the edge. AnomalyMatch took just two and a half days to go through the dataset, which is much faster than if a human research team had attempted this task.

results, published In the magazine Astronomy and astrophysicsIt revealed nearly 1,400 “anomalous objects,” most of which were merging or interacting galaxies. Other anomalies included gravitational lensing (light distorted into circles or arcs by massive objects in front of it), jellyfish galaxies (which have dangling “tentacles” of gas), and galaxies containing large clumps of stars. “Perhaps what is most interesting is the presence of dozens of objects that completely defy classification,” the European Space Agency said in a report. Blog post.

“This is a wonderful use of artificial intelligence to maximize the scientific output of the Hubble Archive,” Gomez said. “Finding so many anomalous objects in the Hubble data, as you might expect to find a lot of them already, is a great result. It also shows how useful this tool is for other large data sets.”