Edinburgh Research Explorer

Scalable Extreme Deconvolution

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publicationProceedings of the Second Workshop on Machine Learning and the Physical Sciences (NeurIPS 2019)
PublisherNeural Information Processing Systems
Number of pages7
Publication statusAccepted/In press - 2 Oct 2019
EventSecond Workshop on Machine Learning and the Physical Sciences (NeurIPS 2019) - Vancouver, Canada
Duration: 14 Dec 201914 Dec 2019
https://ml4physicalsciences.github.io/

Conference

ConferenceSecond Workshop on Machine Learning and the Physical Sciences (NeurIPS 2019)
CountryCanada
CityVancouver
Period14/12/1914/12/19
Internet address

Abstract

The Extreme Deconvolution method fits a probability density to a dataset where each observation has Gaussian noise added with a known sample-specific covariance, originally intended for use with astronomical datasets. The existing fitting method is batch EM, which would not normally be applied to large datasets such as the Gaia catalog containing noisy observations of a billion stars. We propose two minibatch variants of extreme deconvolution, based on an online variation of the EM algorithm, and direct gradient-based optimisation of the log-likelihood, both of which can run on GPUs. We demonstrate that these methods provide faster fitting, whilst being able to scale to much larger models for use with larger datasets.

Event

Second Workshop on Machine Learning and the Physical Sciences (NeurIPS 2019)

14/12/1914/12/19

Vancouver, Canada

Event: Conference

ID: 118415257