Galaxy zoo: Probabilistic morphology through Bayesian CNNs and active learning
Monthly Notices of the Royal Astronomical Society Oxford University Press 491:2 (2019) 1554-1574
Abstract:We use Bayesian convolutional neural networks and a novel generative model of Galaxy Zoo volunteer responses to infer posteriors for the visual morphology of galaxies. Bayesian CNN can learn from galaxy images with uncertain labels and then, for previously unlabelled galaxies, predict the probability of each possible label. Our posteriors are well-calibrated (e.g. for predicting bars, we achieve coverage errors of 11.8 per cent within a vote fraction deviation of 0.2) and hence are reliable for practical use. Further, using our posteriors, we apply the active learning strategy BALD to request volunteer responses for the subset of galaxies which, if labelled, would be most informative for training our network. We show that training our Bayesian CNNs using active learning requires up to 35–60 per cent fewer labelled galaxies, depending on the morphological feature being classified. By combining human and machine intelligence, Galaxy zoo will be able to classify surveys of any conceivable scale on a time-scale of weeks, providing massive and detailed morphology catalogues to support research into galaxy evolution.
A transient search using combined human and machine classifications
MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY 472:2 (2017) 1315-1323
Planet Hunters TESS II: Findings from the first two years of TESS
Monthly Notices of the Royal Astronomical Society 501:4 (2021) 4669-4690
Abstract:© 2021 2020 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society. We present the results from the first two years of the Planet Hunters TESS (PHT) citizen science project, which identifies planet candidates in the TESS (Transiting Exoplanet Survey Satellite) data by engaging members of the general public. Over 22 000 citizen scientists from around the world visually inspected the first 26 sectors of TESS data in order to help identify transit-like signals. We use a clustering algorithm to combine these classifications into a ranked list of events for each sector, the top 500 of which are then visually vetted by the science team. We assess the detection efficiency of this methodology by comparing our results to the list of TESS Objects of Interest (TOIs) and show that we recover 85 per cent of the TOIs with radii greater than 4 R and 51 per cent of those with radii between 3 and 4 R. Additionally, we present our 90 most promising planet candidates that had not previously been identified by other teams, 73 of which exhibit only a single-transit event in the TESS light curve, and outline our efforts to follow these candidates up using ground-based observatories. Finally, we present noteworthy stellar systems that were identified through the Planet Hunters TESS project.
Supermassive black holes in disk-dominated galaxies outgrow their bulges and co-evolve with their host galaxies
Monthly Notices of the Royal Astronomical Society Oxford University Press 470:2 (2017) 1559-1569
Abstract:The deep connection between galaxies and their supermassive black holes is central to modern astrophysics and cosmology. The observed correlation between galaxy and black hole mass is usually attributed to the contribution of major mergers to both. We make use of a sample of galaxies whose disk-dominated morphologies indicate a major-merger-free history and show that such systems are capable of growing supermassive black holes at rates similar to quasars. Comparing black hole masses to conservative upper limits on bulge masses, we show that the black holes in the sample are typically larger than expected if processes creating bulges are also the primary driver of black hole growth. The same relation between black hole and total stellar mass of the galaxy is found for the merger-free sample as for a sample which has experienced substantial mergers, indicating that major mergers do not play a significant role in controlling the coevolution of galaxies and black holes. We suggest that more fundamental processes which contribute to galaxy assembly are also responsible for black hole growth.
Deep learning for automatic segmentation of the nuclear envelope in electron microscopy data, trained with volunteer segmentations
Traffic Wiley 22:7 (2021) 240-253