BEGIN:VCALENDAR VERSION:2.0 PRODID:-//pretalx//pretalx.com//adass2023//talk//9GJPBS BEGIN:VTIMEZONE TZID:MST BEGIN:STANDARD DTSTART:20000101T000000 RRULE:FREQ=YEARLY;BYMONTH=1 TZNAME:MST TZOFFSETFROM:-0700 TZOFFSETTO:-0700 END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:pretalx-adass2023-9GJPBS@pretalx.com DTSTART;TZID=MST:20231106T083000 DTEND;TZID=MST:20231106T083000 DESCRIPTION:AGILE is a space mission launched in 2007 to study X-ray and ga mma-ray astronomy. The AGILE Team is developing new detection algorithms f or Gamma-Ray Bursts (GRBs) both with classical and machine learning techno logies. To train or test these algorithms\, it is necessary to have a larg e GRB dataset\, but usually\, there are not enough real data available. Th is problem is also common for the new generation of high-energy astrophysi cs projects (such as COSI and CTA). It therefore becomes essential to have a system to simulate GRB data. \nThis work aims to develop a Deep Learnin g-based model for generating synthetic GRBs that closely replicate the pro perties and distribution of real GRBs. The dataset obtained using the trai ned model can then be used to develop detection algorithms both with class ic techniques and with machine learning. To develop this generative model we have to take into account several complexities. The main ones are the h uge different temporal behaviours of such GRBs and the lack of data making the training harder.\nWe propose a new method for generating GRBs. This m odel combines the Generative Adversarial Network (GAN) and Variational Aut oencoder (VAE) to produce high-quality and structured generative results w hile preserving latent space structure. We also developed a loss function tailored to our dataset for the reconstruction task. To create the trainin g dataset we extracted the light curves (LC) of GRBs presented in the Four th Fermi-GBM Catalog using only long GRBs. This catalogue contains more th an a decade of observations with a total of 3608 GRBs\, captured through 1 2 NaI and 2 BGO detectors. In this work\, we considered only the NaI dete ctors. We used the LCs detected by multiple sodium detectors and related t o the same GRB as independent\, to augment the number of samples. We filte red the LCs by removing outliers and those with missing values\, ending wi th a total of 5964 LCs. Evaluating the distribution of the GRB duration (t hrough the t90 parameter)\, we set the length of the time series at 220.\n We assessed the model's performance by quantifying the dissimilarity betwe en the histograms of count rates in synthetic and real GRBs. Additionally\ , we conducted quantitative analysis\, employing statistics of the LCs dis tribution in both datasets. The results show that the synthetic LCs are ge nerated with very similar properties to the real ones. We are working on a conditional version of our model using physical parameters of the GRBs s uch as the t90 or the fluence to have a more precise generation. This meth od can be used to generate synthetic LCs for other high-energy astronomy p rojects such as AGILE\, CTA and COSI. DTSTAMP:20260306T060913Z LOCATION:Posters SUMMARY:A new Deep Learning Model for Gamma-ray bursts’ light curves simu lation - Luca Castaldini\, Riccardo Falco URL:https://pretalx.com/adass2023/talk/9GJPBS/ END:VEVENT END:VCALENDAR