2026/07/20 by Zhengyang Zhang, Shuai Zha, Nikhil Sarin +5
Physics and Astronomy · #Gamma-ray bursts and supernovae #Astro and Planetary Science #Neutrino Physics Research
paper · pdf · doi:10.1103/hsms-7v8x36
To address the computational bottleneck of analyzing type II supernova samples from surveys such as the Legacy Survey of Space and Time, we present two STELLA-based neural-network surrogates: an interaction model for low-energy explosions with possible circumstellar-material (CSM) interaction and a photospheric model for standard interaction-free SNe IIP. Each uses an autoencoder to compress spectral energy distributions and an emulator to map physical parameters to the latent space. Latent-mixup regularization improves latent-space continuity, with ResNet blocks used for the interaction model and 2D CNNs for the photospheric model. Their normalized test-set reconstruction MSEs are approximately 9.1e-5 and 1.0e-4, respectively. Applied to SN 2005cs, the interaction model favors a low-mass progenitor, MZAMS = 10.40(+0.04/-0.05) Msun, and confined dense CSM, providing a scenario consistent with direct imaging and helping resolve the historical mass discrepancy. For SN 2012aw, it recovers MZAMS = 11.05(+0.06/-0.06) Msun, consistent with previous studies. For SN 1999em, the photospheric model gives MZAMS = 10.05(+0.07/-0.04) Msun, broadly consistent with preexplosion imaging limits without explicit CSM modeling. These surrogates reduce full Bayesian inference from days to minutes and enable rapid physical characterization of large supernova samples.