http://arxiv.org/abs/2005.07089
In this work, we present a new method to estimate cosmological parameters accurately based on Artificial Neural Network (ANN), and a code called ECoPANN (Estimating Cosmological Parameters with ANN) is developed to achieve parameter inference. We test the ANN method by estimating the basic parameters of the concordance cosmological model using the simulated temperature power spectrum of the cosmic microwave background (CMB). The results show that the ANN performs excellently on best-fit values and errors of parameters, as well as correlations between parameters when compared with that of the Markov chain Monte Carlo (MCMC) method. Besides, for a well trained ANN model, it is capable of estimating parameters for multiple experiments that have different precisions, which can greatly reduce the consumption of time and computing resources for parameter inference. Moreover, the well trained ANN is capable of discovering new potential physics in both the current and future higher precision observations. In addition, we extend the ANN to a multi-branch network to achieve joint constraint on parameters. We test the multi-branch network using the simulated temperature and polarization power spectra of CMB, type Ia supernovae, and baryon acoustic oscillation, and almost obtain the same results as the MCMC method. Therefore, we propose that the ANN can provide an alternative way to accurate and fast estimate cosmological parameters, and ECoPANN can be applied to the research of cosmology and even other broader scientific fields.
G. Wang, S. Li and J. Xia
Fri, 15 May 20
57/65
Comments: 21 pages, 12 figures, and 7 tables. Resubmitted to ApJS after revision. The code repository is available at this https URL