Machine Learning for LSST-scale Citizen Science
Lead Research Organisation:
University of Oxford
Department Name: Oxford Physics
Abstract
Citizen Science - the involvement of volunteers in analysing data - has proved an effective strategy in projects as diverse as galaxy morphology, the search for exoplanets and transient detection and classification. This project involves working directly with the team of web developers and scientists responsible for the world-leading (and Oxford led) Zooniverse.org platform to devise new strategies for effective citizen science in the run up to first light for the Large Synoptic Survey Telescope.
With millions of transient alerts expected each night during the survey's operation, any effective search will require the combination of machine learning with human classifications. This project will focus on methods for detecting and following up on the most unusual objects and events, deploying advanced machine learning techniques for identifying interesting outliers.
Using datasets from existing surveys, the student attached to this project might test and develop appropriate computer vision routines, explore the design of citizen science projects which incorporate their results, or investigate novel modes of combining both. I would also expect a student to contribute to the scientific exploitation of the results from these projects, which will be either related to galaxy morphology or supernovae and other transients.
With millions of transient alerts expected each night during the survey's operation, any effective search will require the combination of machine learning with human classifications. This project will focus on methods for detecting and following up on the most unusual objects and events, deploying advanced machine learning techniques for identifying interesting outliers.
Using datasets from existing surveys, the student attached to this project might test and develop appropriate computer vision routines, explore the design of citizen science projects which incorporate their results, or investigate novel modes of combining both. I would also expect a student to contribute to the scientific exploitation of the results from these projects, which will be either related to galaxy morphology or supernovae and other transients.
Organisations
People |
ORCID iD |
Christopher Lintott (Primary Supervisor) | |
Mike Walmsley (Student) |
Publications
Walmsley M
(2019)
Identification of low surface brightness tidal features in galaxies using convolutional neural networks
in Monthly Notices of the Royal Astronomical Society
Walmsley M
(2020)
Galaxy Zoo: probabilistic morphology through Bayesian CNNs and active learning
in Monthly Notices of the Royal Astronomical Society
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
ST/N504233/1 | 30/09/2015 | 30/03/2021 | |||
1947724 | Studentship | ST/N504233/1 | 30/09/2017 | 30/03/2021 | Mike Walmsley |
ST/R505006/1 | 30/09/2017 | 29/09/2021 | |||
1947724 | Studentship | ST/R505006/1 | 30/09/2017 | 30/03/2021 | Mike Walmsley |
Title | Deep learning classifier for galaxy morphology (prototype) |
Description | Deep learning model that can predict, with uncertainty, how a human would describe a galaxy |
Type Of Technology | Software |
Year Produced | 2019 |
Open Source License? | Yes |
Impact | Will support future large catalogs of galaxy shapes |
Description | East London School Visit - Christ Church Horizons |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Schools |
Results and Impact | Taster astrophysics and Oxford application workshop for school in Barnet, London |
Year(s) Of Engagement Activity | 2019 |
Description | Stargazing Live 2018, 2019 |
Form Of Engagement Activity | Participation in an open day or visit at my research institution |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Public/other audiences |
Results and Impact | Share Zooniverse website and research as part of annual departmental outreach day |
Year(s) Of Engagement Activity | 2018,2019 |
URL | https://www2.physics.ox.ac.uk/events/2019/01/26/stargazing-oxford-2019 |