Orateur
Description
Inverse problems in astronomy are often computationally expensive, and Markov Chain Monte Carlo (MCMC) methods become impractical when dealing with massive, heterogeneous datasets or when only limited observations are available. With the advent of the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST), the astronomical community is expected to receive millions of transient alerts per night, many of which evolve on very short timescales. Rapid classification and characterization of these events is therefore essential.
In this work, we present a Physics-Informed Neural Network (PINN) framework to model the historical light curves of Fink alerts. Using the Type II-P supernova SN~2022acko as a test case, we demonstrate that our method can accurately infer the physical properties of the transient progenitor system, achieving reliable results within only a few days of observations. In contrast to traditional MCMC approaches, this method requires minimal human intervention, enabling scalable, automated, and physics-driven characterization of large volumes of transients in the Fink alert stream.