First Supernova Discovered by AI
A new AI tool has helped to successfully detect, identify, and classify a supernova in a first for AI. Called Bright Transient Survey BO or BTSbot for short, is a product of an international collaboration led by Northwestern University, University of Minnesota, Liverpool John Mores University in England, and Stockholm University in Sweden.
In the university’s statement, they showcased that this was the first instance of an AI being able to classify a new supernova candidate while bypassing human error. If proven to be used on a wide scale, it could remove humans from the overall process of identifying supernovas and reduce false positives in the process.
The team who came up with the AI, informed the astronomical community of the launch and success of the tool this past week. According to them, humans have spent over twenty-two hundred hours visually inspecting and classifying supernova candidates.
But with BTSbot, required man-hours can be scaled back and redirected to other areas of research. Freeing up teams to be allowed to accelerate their work. This has been one major aspect of AI that has been seen across fields and disciplines; its ability to reduce required human-hour labor to optimize workflows.
Adam Miller of Northwestern who led the work said of the project in a statement, “For the first time, a series of robots and AI algorithms has observed, then identified, then communicated with another telescope to finally confirm the discovery of a supernova.”
Miller continued, “This represents an important step forward as further refinement of models will allow the robots to isolate specific subtypes of stellar explosions. Ultimately, removing humans from the loop provides more time for the research team to analyze their observations and develop new hypotheses to explain the origin of the cosmic explosions that we observe.”
Nabeel Rehemtulla who co-led BTSbot with Miller spoke of how the AI could enhance astronomical work, “This significantly streamlines large studies of supernovae, helping us better understand the life cycles of stars and the origin of elements supernovae create, like carbon, iron and gold.”
Basically what the AI does is that it removes the naked eye confirmation of visual data. Miller provides insights into the process. “Automated software presents a list of candidate explosions to humans, who spend time verifying the candidates and executing spectroscopic observations.”
Continuing, “We can only definitively know that a candidate is truly a supernova by collecting its spectrum — the source’s dispersed light, which reveals elements present in the explosion. There are existing robotic telescopes that can collect spectra, but this is also often done by humans operating telescopes with spectrographs.”
The team trained BTSbot on over 1.4 million historical images from nearly 16,000 sources. These included confirmed supernovae, temporary flaring stars and galaxies, and other visual data. By providing this to the AI, it allowed it to learn how to both spot a supernova and spot what’s likely a false positive.
Nabeel Rehemtulla elegantly described how it felt seeing the BTSbot work, “The beauty of it is that, once everything is turned on and working properly, we don’t do anything. We go to sleep at night, and, in the morning, we see that BTSbot and these other AIs unwaveringly do their jobs.”
This isn’t the first case of AI entering the realm of Astronomy. AI is also helping researchers identify gamma-ray bursts, the universe’s most powerful and violent expositions.
Originally posted on OpenDataScience.com
Read more data science articles on OpenDataScience.com, including tutorials and guides from beginner to advanced levels! Subscribe to our weekly newsletter here and receive the latest news every Thursday. You can also get data science training on-demand wherever you are with our Ai+ Training platform. Interested in attending an ODSC event? Learn more about our upcoming events here.
