Predicting galaxy spectra from images with hybrid convolutional neural networks
This work addresses the challenge of obtaining spectroscopic data for galaxies, benefiting astronomers by enabling more efficient analysis for future wide-field surveys like the Vera C. Rubin Observatory.
The authors tackled the problem of predicting galaxy spectra from broad-band imaging, which is observationally expensive, and achieved robust predictions for the first time using a hybrid convolutional neural network that outperformed other models.
Galaxies can be described by features of their optical spectra such as oxygen emission lines, or morphological features such as spiral arms. Although spectroscopy provides a rich description of the physical processes that govern galaxy evolution, spectroscopic data are observationally expensive to obtain. For the first time, we are able to robustly predict galaxy spectra directly from broad-band imaging. We present a powerful new approach using a hybrid convolutional neural network with deconvolution instead of batch normalization; this hybrid CNN outperforms other models in our tests. The learned mapping between galaxy imaging and spectra will be transformative for future wide-field surveys, such as with the Vera C. Rubin Observatory and Nancy Grace Roman Space Telescope, by multiplying the scientific returns for spectroscopically-limited galaxy samples.