Supernova Classification using the Recurrent Neural Network in the CSST Ultra-Deep Field Survey

We study supernova (SN) classification using the Recurrent Neural Networks (RNNs) within the Chinese Space-station Survey Telescope Ultra-Deep Field (CSST-UDF) photometric survey and explore the improvements in cosmological constraints. We simulate Type Ia supernovae (SNe Ia) and core-collapse supernovae (CCSNe) using SNCosmo with SALT3 SN Ia model and CCSN templates, and apply the SuperNNova (SNN) program for classification. Our study indicates that the SNN combined with the Joint Light-curve Analysis cuts can enhance the purity of the CSST-UDF SN Ia sample up to over 99.5% with 2,193 SNe Ia and 4 CCSNe, which can significantly increase the reliability of the cosmological constraints. The method based on the Bayesian Estimation Applied to Multiple Species with Bias Corrections framework is used to correct the SN Ia magnitude bias caused by the selection effect and CCSN contamination, and the Markov Chain Monte Carlo (MCMC) method is employed for cosmological constraints. We find that the accuracy of the constraints on the matter density $\Omega_{\rm M}$ and the equation of state of dark energy parameter $w$ can achieve 14% and 18%, respectively, assuming the flat $w$CDM model. This result is comparable to current surveys relying on spectroscopic confirmation. Our results indicate that our data analysis method is effective, and the CSST-UDF SN photometric survey is a powerful tool to explore the expansion history of the Universe.