In Hyderabad, advancements in Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing astronomical research, particularly in the identification of supernovae—rare stellar explosions marking the end of massive stars. Researchers from Oxford University and Queen’s University Belfast utilize the Asteroid Terrestrial Impact Last Alert System (ATLAS), funded by NASA, to scan the visible sky every 24 to 48 hours. This system generates millions of alerts nightly, which are filtered down to 200-400 potential supernova signals requiring manual review.
To streamline this process, Oxford developed the Virtual Research Assistant (VRA), an AI tool that automates much of the data analysis. By using a smaller dataset—15,000 examples—the VRA significantly reduces astronomers’ workload by 85%, while maintaining high accuracy. In its first year, it processed over 30,000 alerts, missing fewer than 0.08% of real supernovae. As the Vera Rubin Observatory’s Legacy Survey of Space and Time prepares for its launch, the VRA’s contributions are increasingly vital for astronomical discoveries.
Source link
