Domenico Sapone, Universidad de Chile, FCFM
Resumen:
We propose a novel approach using neural networks (NNs) to differentiate between cosmological models, implementing LIME as an interpretability tool to identify the key features driving the model’s decisions. We demonstrate the potential of NNs to enhance the extraction of meaningful information from large-scale structure data, based on current galaxy-clustering survey specifications, for both the standard cosmological model and the Hu–Sawicki model. Our results show that the NN can successfully distinguish between the two models, achieving an overall accuracy of approximately 97%. This highlights the potential of NNs to maximize the scientific return of current and next-generation surveys in probing deviations from general relativity.
La charla se realizará el miércoles 17 de junio, de 16:20 a 17:50 hrs, en la sala A2FM (Edificio A, campus Fernando May), y contará con un coffee break.
