ASPCS
 
Back to Volume
Paper: Prediction of molecular parameters from astronomical emission lines, using Neural Networks
Volume: 535, Astronomical Data Analysis Software and Systems XXXI
Page: 107
Authors: Barrientos, A.; Solar, M.
Abstract: We investigate if neural networks can be used to predict excitation temperature (Tex) and column density (log(N)) for a given spectrum. We used as test cases the spectra of CO, HCO+, SiO and CH3CN between 80 and 400 GHz. Training spectra were generated with MADCUBA, a state-of-the-art spectral analysis tool. Our algorithm was designed to allow the generation of predictions for multiple molecules in parallel, in a way that is scalable, and presents a linear speedup. We were able to predict the column density and excitation temperature of these molecules with a mean absolute error of 8.5% for CO, 4.1% for HCO+, 1.5% for SiO and 1.6% for CH3CN.
Back to Volume