Learning the Relationship between Galaxies Spectra and their Star Formation Histories using Convolutional Neural Networks and Cosmological Simulations

Abstract: We present a new method for inferring galaxy star formation histories (SFH) using machine learning methods coupled with two cosmological hydrodynamic simulations. We train Convolutional Neural Networks to learn the relationship between synthetic galaxy spectra and high resolution SFHs from the EAGLE and Illustris models. To evaluate our SFH reconstruction we use Symmetric Mean Absolute Percentage Error (SMAPE), which acts as a true percentage error in the low-error regime. On dust-attenuated spectra we achieve high test accuracy (median SMAPE $= 10.5\%$). Including the effects of simulated observational noise increases the error ($12.5\%$), however this is alleviated by including multiple realisations of the noise, which increases the training set size and reduces overfitting ($10.9\%$). We also make estimates for the observational and modelling errors. To further evaluate the generalisation properties we apply models trained on one simulation to spectra from the other, which leads to only a small increase in the error (median SMAPE $\sim 15\%$). We apply each trained model to SDSS DR7 spectra, and find smoother histories than in the VESPA catalogue. This new approach complements the results of existing SED fitting techniques, providing star formation histories directly motivated by the results of the latest cosmological simulations.