The Euclid telescope, due for launch in 2021, will perform an imaging and slitless spectroscopy survey over half the sky, to map baryon wiggles and weak lensing. During the survey Euclid is expected to resolve 100,000 strong gravitational lens systems. This is ideal to find rare lens configurations, provided they can be identified reliably and on a reasonable timescale. For this reason we have developed a Convolutional Neural Network (CNN) that can be used to identify images containing lensing systems. CNNs have already been used for image and digit classification as well as being used in astronomy for star-galaxy classification. Here our CNN is trained and tested on Euclid-like and KiDS-like simulations from the Euclid Strong Lensing Group, successfully classifying 77% of lenses, with an area under the ROC curve of up to 0.96. Our CNN also attempts to classify the lenses in COSMOS HST F814W-band images. After convolution to the Euclid resolution, we find we can recover most systems that are identifiable by eye. The Python code is available on Github.

We investigate the constraints on models of supermassive black hole (SMBH) and quasar formation obtainable from two recent observational developments: the discovery of luminous quasars at z~6, and estimates of the local mass density of SMBHs. If ~90 per cent of this mass was accreted at redshifts z<3, as suggested by the observed quasar luminosity functions, these joint constraints pose a challenge for models, which must account for the observed luminous quasar population at z~6 within a very limited `mass budget'. We investigate a class of models based within the hierarchical structure formation scenario, in which major mergers lead to black hole formation and fuelling, and the resulting quasars shine at their Eddington-limited rate until their fuel is exhausted. We show that the simplest such model, in which a constant fraction of the gas within the halo is accreted in each major merger, cannot satisfy both constraints simultaneously. When this model is normalized to reproduce the number density of luminous quasars at z~6, the mass budget is grossly exceeded due to an overabundance of lower mass SMBHs. We explore a range of modifications to the simple model designed to overcome this problem. We show that both constraints can be satisfied if the gas accretion fraction scales as a function of the halo virial velocity. Similar scalings have been proposed in order to reproduce the local M-sigma relation. Successful models can also be constructed by restricting the formation of seed black holes to redshifts above z~11.5 or to haloes above a velocity threshold ~55 km/s, or assuming that only a fraction of major mergers result in formation of a seed SMBH. (abridged)