Geological mapping in Mercury's southern latitudes

Description:
The aim of this project is to use a relatively new machine learning technology, Generative
Adversarial Networks, to predict the unresolved structures in submillimetre-wavelength
images and then use these predictions to find candidate ultra-high redshift galaxies.
Extreme high-redshift galaxies can be detected in submm-wavelength surveys made by the
Herschel Space Observatory and others, but because of the redshifting, these galaxies are
typically only detected at the longest wavelengths, e.g. 500 microns. At these wavelengths,
the angular resolution is at its coarsest, so distant galaxies are seen as blended with
foreground interlopers. Several groups have attempted to get around this by modelling the
data at all wavelengths. This approach deconvolves the longest wavelength data by using
the locations of galaxies detected at other wavelengths as a prior, but this is
computationally difficult and solutions are not unique. Generative Adversarial Networks are
a relatively new machine learning technology that, once trained, can reconstruct missing
information in images. This has been shown to work in blurred images of galaxies,
accurately reconstructing galaxy morphologies in many cases. This project will deploy this
technology on real and simulated submm-wave images, find candidate ultra-high redshift
galaxies, follow them up with ground-based and space observatories, place statistical
constraints on ultraluminous star forming galaxies at the earliest accessible epochs and
constrain models of high redshift star formation