Hamelijnck, Oliver, Damoulas, Theodoros, Wang, Kangrui and Girolami, Mark (2019) Multi-resolution multi-task Gaussian processes. In: 33rd Conference on Neural Information Processing Systems, Canada, 8-14 Dec 2019. Published in: 33nd Conference on Neural Information Processing Systems (NeurIPS 2019), 32 pp. 14048-14058.

Preview

PDF WRAP-multi-resolution-multi-task-Gaussian-processes-Damoulas-2019.pdf - Accepted Version - Requires a PDF viewer. Download (1234Kb) | Preview

Request Changes to record.

Abstract

We consider evidence integration from potentially dependent observation processes under varying spatio-temporal sampling resolutions and noise levels. We develop a multi-resolution multi-task (MRGP) framework while allowing for both inter-task and intra-task multi-resolution and multi-fidelity. We develop shallow Gaussian Process (GP) mixtures that approximate the difficult to estimate joint likelihood with a composite one and deep GP constructions that naturally handle biases in the mean. By doing so, we generalize and outperform state of the art GP compositions and offer information-theoretic corrections and efficient variational approximations. We demonstrate the competitiveness of MRGPs on synthetic settings and on the challenging problem of hyper-local estimation of air pollution levels across London from multiple sensing modalities operating at disparate spatio-temporal resolutions.

Item Type:	Conference Item (Paper)
Subjects:	Q Science > QA Mathematics T Technology > TD Environmental technology. Sanitary engineering T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Faculty of Science, Engineering and Medicine > Science > Computer Science
Library of Congress Subject Headings (LCSH):	Sensor networks -- Data processing, Multisensor data fusion, Gaussian processes, Air -- Pollution -- England -- London -- Data processing
Series Name:	Advances in Neural Information Processing Systems
Journal or Publication Title:	33nd Conference on Neural Information Processing Systems (NeurIPS 2019)
Publisher:	Curran Associates, Inc
Official Date:	December 2019
Dates:	Date Event December 2019 Published 11 December 2019 Completion 4 September 2019 Accepted
Volume:	32
Page Range:	pp. 14048-14058
Article Number:	1258
Status:	Peer Reviewed
Publication Status:	Published
Access rights to Published version:	Restricted or Subscription Access
Date of first compliant deposit:	12 September 2019
Date of first compliant Open Access:	16 February 2022
RIOXX Funder/Project Grant:	Project/Grant ID RIOXX Funder Name Funder ID London Air Quality project Lloyd's Register Foundation http://dx.doi.org/10.13039/100008885 UNSPECIFIED Alan Turing Institute http://dx.doi.org/10.13039/100012338 EP/N510129/1 [EPSRC] Engineering and Physical Sciences Research Council http://dx.doi.org/10.13039/501100000266 UNSPECIFIED Greater London Authority http://dx.doi.org/10.13039/100013998
Conference Paper Type:	Paper
Title of Event:	33rd Conference on Neural Information Processing Systems
Type of Event:	Conference
Location of Event:	Canada
Date(s) of Event:	8-14 Dec 2019
Related URLs:	Organisation Open Access File
Open Access Version:	ArXiv Publisher

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads