Statistical abstraction for multi-scale spatio-temporal systems

Michalis Michaelides, Jane Hillston, Guido Sanguinetti

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


Spatio-temporal systems exhibiting multi-scale behaviour are common in applications ranging from cyber-physical systems to systems biology, yet they present formidable challenges for computational modelling and analysis. Here we consider a prototypic scenario where spatially distributed agents decide their movement based on external inputs and a fast-equilibrating internal computation. We propose a generally applicable strategy based on statistically abstracting the internal system using Gaussian Processes, a powerful class of non-parametric regression techniques from Bayesian Machine Learning. We show on a running example of bacterial chemotaxis that this approach leads to accurate and much faster simulations in a variety of scenarios.
Original languageEnglish
Title of host publicationInternational Conference on Quantitative Evaluation of Systems QEST 2017
Subtitle of host publicationQuantitative Evaluation of Systems
PublisherSpringer, Cham
Number of pages16
ISBN (Electronic)978-3-319-66335-7
ISBN (Print)978-3-319-66334-0
Publication statusPublished - 11 Aug 2017
Event14th International Conference on Quantitative Evaluation of Systems - Berlin, Germany
Duration: 5 Sep 20177 Sep 2017

Publication series

NameLecture Notes in Computer Science
Publisher Springer, Cham
ISSN (Print)0302-9743


Conference14th International Conference on Quantitative Evaluation of Systems
Abbreviated titleQEST 2017
Internet address


Dive into the research topics of 'Statistical abstraction for multi-scale spatio-temporal systems'. Together they form a unique fingerprint.

Cite this