Galaxy cluster mass estimation with deep learning and hydrodynamical simulations

We evaluate the ability of Convolutional Neural Networks (CNNs) to predict
galaxy cluster masses in the BAHAMAS hydrodynamical simulations. We train
four separate single-channel networks using: stellar mass, soft X-ray
flux, bolometric X-ray flux, and the Compton y parameter as observational
tracers, respectively.  Our training set consists of ~6400 synthetic
cluster images generated from the simulation, while an additional ~1600
images form a test set.  We also train a "multi-channel" CNN by combining
the four observational tracers.  The cluster masses predicted from these
networks are evaluated using the average fractional difference between
predicted cluster mass and true cluster mass.  The resulting predictions
are especially precise for halo masses in the range
10^13.25 to 10^14.5 MSun, where all five networks
produce mean mass biases of order ~1% with a scatter on the mean bias of
~0.5%. The network trained with Compton y parameter maps yields the most
precise predictions. We interpret the network's behaviour using two
diagnostic tests to determine which features are used to predict cluster
mass. The CNN trained with stellar mass images detect galaxies (not
surprisingly), while CNNs trained with gas-based tracers utilise the shape
of the signal to estimate cluster mass.