The ASTRAI project focuses on harnessing Data Science (GenAI and other Machine Learning) to enhance our understanding of supernovae (SNe), which are vital to numerous astrophysical processes. SNe play a crucial role in various astrophysical processes, including nucleosynthesis, dust production, cosmic ray acceleration, and the emission of neutrinos and gravitational waves.

The ASTRAI project has been selected and funded under the ICSC Programme, financed as part of the PNRR, funded by the European Union – NextGenerationEU.

Today, scientists study SNe using complex statistical methods like Gaussian Processes to reconstruct light and velocity curves. This process is time-consuming, often taking hours for a single curve, which slows down the overall pace of scientific research. Moreover, the upcoming Legacy Survey of Space and Time (LSST) is expected to generate a massive amount of observational data that will need to be analysed quickly.

To address these challenges, ASTRAI proposes the use of Machine Learning (ML) techniques. While ML holds great promise, it requires large training datasets. Currently, there are no sufficiently large datasets of SNe available. To overcome this limitation, ASTRAI will use Generative AI techniques to create a synthetic dataset. The large synthetic dataset can then be used as a training dataset for an automatic data analysis system, significantly accelerating the pace of SNe research.