Seven possible quasar lenses emerge from AI scan of 800,000 DESI objects

Quasars, distant cores of galaxies powered by supermassive black holes, are among the most luminous objects in the universe. While not uncommon, their brightness can make it difficult to accurately measure the galaxies they reside in. This means scientists must use gravitational lensing to assist in analyzing these bright objects, a method that relies on studying how an object’s strong gravity bends light around its host galaxy. Yet despite their own powerful gravity, quasars that can act as lenses are rare.

Moreover, while nearly every galaxy is home to a black hole, research suggests those that form quasars may act as “missing links” in understanding the formation and evolution of the early universe. Young ones, especially, could be key to unlocking vast cosmic secrets.

Now, to identify more quasars as gravitational lenses, an international team of researchers analyzed a list of 800,000 quasars from the Dark Energy Spectroscopic Instrument (DESI) survey. Using an AI model trained on a small sample of mock lenses, or fake examples of quasar lens systems, to automatically search for these rare events, researchers found seven new candidates.

“Quasars are like the baby pictures of a supermassive black hole,” said Everett McArthur, lead author of the study and a graduate student in astronomy at The Ohio State University. “So exploring how we get from quasars to those black holes is really important.”

These new candidates double the number of quasars scientists found in surveys in years past. With more data, the discovery offers an opportunity to expand knowledge of how their systems work, as well as how the galaxies they reside in grow and evolve.

For instance, while the seven candidates in this study are located at least 5–6 billion light-years from Earth, uncovering new insights about these faraway objects could also reveal valuable information about our own galaxy, McArthur said.

“By studying the tight correlation between galaxies and black holes, we could understand why our galaxy is the way that it is and perhaps why our own black hole is sometimes dormant,” he said.

The study was published in The Astrophysical Journal.

Teaching AI to spot rare lenses

Beyond the team’s observations, what is unique about its work is its use of neural networks. Since there aren’t enough real-life examples of quasars acting as lenses, researchers had to teach their AI to identify the emission lines of potential quasars acting as gravitational lenses using a mixture of real quasar and background galaxy spectra.

This method created a simulation so impressive that the AI was able to recognize the subtle differences between normal and abnormal quasars with unique features, McArthur said.

“What this proves is our architecture was able to parse through a diverse array of quasar spectra in a really significant way,” McArthur said.

After whittling DESI’s list of 800,000 potential quasars to 200, the team hand-reviewed the shortened list before narrowing the candidates to a final seven.

From seven candidates to confirmation

Going forward, the researchers will seek to directly confirm their observations using powerful space-based instruments like the Hubble Space Telescope. Once those deeper studies are completed, they expect to use their AI model to help future scientists search for and validate other kinds of strange cosmic phenomena.

“You can very well expand this type of study to find many rare anomalies in a spectrum,” McArthur said. “We’re in an era when science has suddenly become more accessible than ever, and applying AI to astronomy and machine-learning methods to big data sets is part of that.”

Who’s behind this story?

Lisa Lock

BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021.

Full profile →

Robert Egan

Bachelor’s in mathematical biology, Master’s in creative writing. Well-traveled with unique perspectives on science and language.

Full profile →