Scientists at the California Institute of Technology used machine learning to classify 1,000 supernovae caused by the explosions of dying stars. Writes about it space.com.
Astronomers using the SNIascore algorithm created a catalog based on data collected by the Zwicky Transient Facility (ZTF) instrument attached to the Samuel Oschin telescope at the institute’s Palomar Observatory.
According to scientists, ZTF scans the night sky and collects a huge amount of data that is problematic to process manually. To solve this problem, they developed the SNIAscore algorithm.
“We knew that as soon as we trained our computers to do this work, they would take a lot of work off us,” said Christopher Fremling of the California Institute of Technology and one of the authors of the project.
Since 2017, the ZTF has identified thousands of supernovae, which can be divided into two classes:
- hydrogen-free Type I;
- hydrogen-rich Type II.
Most often, a supernova of the first type occurs due to the absorption of matter from a nearby donor star, which falls on its surface and causes a thermonuclear explosion. Type II forms when massive stars run out of fuel needed for nuclear fusion and can no longer withstand gravitational collapse.
SNIascore classifies a special type of supernova called Type Ia. They occur when a dying star explodes and forms a uniform light output called “standard candles”. Astronomers use information about them to measure distances in space and determine the expansion rate of the universe.
Every night after ZTF searches the sky for transient events and objects, artificial intelligence proceeds to classify Type Ia stars.
“SNIascore detected its first supernova in April 2021, and a year and a half later, we reached the milestone of 1,000 objects,” Fremling said.
According to him, the algorithm works almost flawlessly. In the future, scientists plan to implement the same model with other means of observation.
Astronomers also intend to use SNIAscore to classify other types of supernovae. Even before these advances take place, a machine learning tool is changing astronomy and showing the changing face of this scientific field.
Recall that in June 2021, artificial intelligence discovered hidden links between the Milky Way and Andromeda.
In July, scientists presented an algorithm that classifies thousands of galaxies per second.
In December, astronomers cataloged 170 rogue exoplanets discovered by artificial intelligence.
Subscribe to Cryplogger news in Telegram: Cryplogger AI – all the news from the world of AI!
Found a mistake in the text? Select it and press CTRL+ENTER
Scientists at the California Institute of Technology used machine learning to classify 1,000 supernovae caused by the explosions of dying stars. Writes about it space.com.
Astronomers using the SNIascore algorithm created a catalog based on data collected by the Zwicky Transient Facility (ZTF) instrument attached to the Samuel Oschin telescope at the institute’s Palomar Observatory.
According to scientists, ZTF scans the night sky and collects a huge amount of data that is problematic to process manually. To solve this problem, they developed the SNIAscore algorithm.
“We knew that as soon as we trained our computers to do this work, they would take a lot of work off us,” said Christopher Fremling of the California Institute of Technology and one of the authors of the project.
Since 2017, the ZTF has identified thousands of supernovae, which can be divided into two classes:
- hydrogen-free Type I;
- hydrogen-rich Type II.
Most often, a supernova of the first type occurs due to the absorption of matter from a nearby donor star, which falls on its surface and causes a thermonuclear explosion. Type II forms when massive stars run out of fuel needed for nuclear fusion and can no longer withstand gravitational collapse.
SNIascore classifies a special type of supernova called Type Ia. They occur when a dying star explodes and forms a uniform light output called “standard candles”. Astronomers use information about them to measure distances in space and determine the expansion rate of the universe.
Every night after ZTF searches the sky for transient events and objects, artificial intelligence proceeds to classify Type Ia stars.
“SNIascore detected its first supernova in April 2021, and a year and a half later, we reached the milestone of 1,000 objects,” Fremling said.
According to him, the algorithm works almost flawlessly. In the future, scientists plan to implement the same model with other means of observation.
Astronomers also intend to use SNIAscore to classify other types of supernovae. Even before these advances take place, a machine learning tool is changing astronomy and showing the changing face of this scientific field.
Recall that in June 2021, artificial intelligence discovered hidden links between the Milky Way and Andromeda.
In July, scientists presented an algorithm that classifies thousands of galaxies per second.
In December, astronomers cataloged 170 rogue exoplanets discovered by artificial intelligence.
Subscribe to Cryplogger news in Telegram: Cryplogger AI – all the news from the world of AI!
Found a mistake in the text? Select it and press CTRL+ENTER