Cosmic leap: NASA Swift satellite and AI unravel the distance of the farthest gamma-ray bursts

The advent of AI is being hailed by many as a societal game-changer, as it opens up a universe of possibilities to improve virtually every aspect of our lives.

Astronomers are now using AI, quite literally, to measure the expansion of our universe.

Two recent studies led by Maria Dainotti, visiting professor at UNLV’s Nevada Center for Astrophysics and assistant professor at the National Astronomical Observatory of Japan (NAOJ), integrated multiple machine learning models to add a new level of precision to distance measurements for gamma ray bursts (GRBs) – the most luminous and violent explosions in the universe.

In just a few seconds, GRBs release the same amount of energy that our sun releases during its entire life. Because they are so bright, GRBs can be observed at multiple distances – including at the edge of the visible universe – and can help astronomers in their search for the oldest and most distant stars. But due to the limitations of current technology, only a small percentage of known GRBs have all the observational signatures astronomers need to calculate how far away they occurred.

Dainotti and her teams combined GRB data from NASA’s Neil Gehrels Swift Observatory with multiple machine learning models to overcome the limitations of current observation technology and, more precisely, estimate the proximity of GRBs whose distance is unknown. Because GRBs can be observed both far away and at relatively short distances, knowing where they occur can help scientists understand how stars evolve over time and how many GRBs might exist in a given space and time.

“This research pushes the boundaries in both gamma-ray astronomy and machine learning,” says Dainotti. “Further research and innovation will help us achieve even more reliable results and allow us to answer some of the most pressing cosmological questions, including the earliest processes of our universe and how it has evolved over time.”

AI Expands the Limits of Deep Space Observation In one study, Dainotti and Aditya Narendra, a final-year doctoral student at Poland’s Jagiellonian University, used several machine learning methods to accurately measure the distance of GRBs observed by the Swift UltraViolet/Optical Telescope in the space. UVOT) and ground-based telescopes, including the Subaru telescope. The measurements were based solely on other, non-distance-related GRB properties. The research was published on May 23 in the Astrophysical diary letters.

“The outcome of this study is so accurate that using the predicted distance, we can determine the number of GRBs in a given volume and time (called the velocity), which is very close to the actual observed estimates,” said Narendra.

Another study led by Dainotti and international collaborators has been successful in measuring GRB distance with machine learning using data from NASA’s Swift X-ray Telescope (XRT) afterglows of so-called long GRBs. GRBs are believed to occur in a variety of ways. Long GRBs are formed when a massive star reaches the end of its life and explodes in a spectacular supernova. Another type, known as short GRBs, is formed when the remnants of dead stars, such as neutron stars, gravitationally merge and collide with each other.

Dainotti says the novelty of this approach comes from using different machine learning methods together to improve their collective predictive power. This method, called Superlearner, assigns each algorithm a weight whose values ​​range from 0 to 1, with each weight corresponding to the predictive power of that unique method.

“The advantage of the Superlearner is that the final prediction always outperforms the single models,” says Dainotti. “Superlearner is also used to weed out the algorithms that are the least predictive.”

This study, which was published on February 26 in The Astrophysical Journal, Supplement Seriesreliably estimates the distance of 154 long GRBs whose distance is unknown and significantly increases the population of known distances under this type of eruption.

Answering puzzling questions about GRB formation

A third study, published February 21 in the Astrophysical diary letters and led by astrophysicist Vahé Petrosian and Dainotti of Stanford University, Swift used X-ray data to answer puzzling questions by showing that the GRB rate – at least at small relative distances – does not track the rate of star formation.

“This opens the possibility that long GRBs at small distances are not generated by the collapse of massive stars, but rather by the merger of very dense objects such as neutron stars,” says Petrosian.

With support from NASA’s Swift Observatory Guest Investigator program (cycle 19), Dainotti and her colleagues are now working to make the machine learning tools publicly available through an interactive web application.