For long, detecting supernovas, a bright and massive explosion of a star, has depended on humans and computers working together. Now, in a first, a new artificial intelligence (AI) tool has successfully detected, identified and classified a supernova.
The AI tool, Bright Transient Survey Bot (BTSbot), developed by an international collaboration led by Northwestern University, automates the entire search for new supernovae across the night sky, which could effectively remove humans from the process, a press statement by the university explains. Notably, the AI accelerates the process and also bypasses the element of human error.
Many supernovas occur when stars reach the last evolutionary stage and exhaust their fuel for nuclear fusion, a Space.com report explains. When they can no longer support themselves against the inward push of gravity, their cores collapse while their outer layers blast out as supernovas.
The tool’s machine learning algorithm was trained using over 1.4 million images from around 16,000 sources, including confirmed supernovae, temporarily flaring stars, periodically variable stars and flaring galaxies. It was developed to remove humans from the process. In the past six years, humans have spent about 2,200 hours visually inspecting and classifying supernova candidates, the statement adds.
“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,” Adam Miller, who led the study, says in the statement.
In the statement, Nabeel Rehemtulla, who co-led the study, adds that the AI tools significantly streamline large studies of supernovae, and help understand the life cycle of stars as well as the origin of elements supernovae create, like carbon, iron, and gold.
Currently, to detect and analyse supernovae, humans work with robotic systems. First, robotic telescopes repeatedly image the same sections of the night sky, in a quest to detect new sources, the statement explains. When something new is detected, humans take charge. Automated software provides a list of candidate explosions, and humans spend time verifying them.
If the BTSbot is added to this workflow then humans won’t have to spend a significant amount of time inspecting the candidates. Rehemtulla says. “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.”