dc.contributor.author | Kirkwood, C | |
dc.contributor.author | Economou, T | |
dc.contributor.author | Odbert, H | |
dc.contributor.author | Pugeault, N | |
dc.date.accessioned | 2021-03-25T14:43:14Z | |
dc.date.issued | 2021-03-24 | |
dc.description.abstract | Deep learning – machine learning using deep neural networks – is an efficient way to discover patterns in data that may be more complex than we could manually hypothesise. Here we learn spatio-temporal models that harness information from gridded auxiliary datasets, such as digital terrain models and satellite imagery, by learning task-relevant derivatives of these with no requirement for manual feature engineering. By operating within the Bayesian probabilistic framework, we can learn well-calibrated deep models that quantify epistemic and aleatoric uncertainties and avoid overfitting despite the capacity of deep models to do so. | en_GB |
dc.description.sponsorship | Engineering and Physical Sciences Research Council (EPSRC) | en_GB |
dc.identifier.citation | The Turing Presents: AI UK, 23–24 March 2021, The Alan Turing Institute, UK. Online | en_GB |
dc.identifier.uri | http://hdl.handle.net/10871/125229 | |
dc.language.iso | en | en_GB |
dc.publisher | Alan Turing Institute | en_GB |
dc.relation.url | https://www.turing.ac.uk/ai-uk | en_GB |
dc.rights | © 2021 Alan Turing Institute | en_GB |
dc.subject | deep learning | en_GB |
dc.subject | environmental science | en_GB |
dc.subject | geostatistics | en_GB |
dc.subject | bayesian statistics | en_GB |
dc.subject | weather | en_GB |
dc.subject | uncertainty | en_GB |
dc.title | Bayesian deep learning for large scale environmental data modelling | en_GB |
dc.type | Conference paper | en_GB |
dc.date.available | 2021-03-25T14:43:14Z | |
dc.rights.uri | http://www.rioxx.net/licenses/all-rights-reserved | en_GB |
pubs.funder-ackownledgement | Yes | en_GB |
dcterms.dateAccepted | 2021-03-05 | |
exeter.funder | ::Engineering and Physical Sciences Research Council (EPSRC) | en_GB |
rioxxterms.version | VoR | en_GB |
rioxxterms.licenseref.startdate | 2021-03-05 | |
rioxxterms.type | Conference Paper/Proceeding/Abstract | en_GB |
refterms.dateFCD | 2021-03-25T14:40:36Z | |
refterms.versionFCD | VoR | |
refterms.dateFOA | 2021-03-25T14:43:32Z | |
refterms.panel | B | en_GB |