Modeling the Drift Function in Stochastic Differential Equations using Reduced Rank Gaussian Processes

Roland Hostettler, Filip Tronarp, Simo Särkkä

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

2 Citations (Scopus)
157 Downloads (Pure)

Abstract

In this paper, we propose a Gaussian process-based nonlinear, time-varying drift model for stochastic differential equations. In particular, we combine eigenfunction expansion of the Gaussian process’ covariance kernel in the spatial input variables with spectral decomposition in the time domain to obtain a reduced rank state space representation of the drift model, which avoids the growing complexity (with respect to time) of the full Gaussian process solution. The proposed approach is evaluated on two nonlinear benchmark problems, the Bouc Wen and the cascaded tanks systems.

Original languageEnglish
Title of host publication18th IFAC Symposium on System Identification, SYSID 2018
PublisherElsevier
Pages778-783
Number of pages6
Volume51
Edition15
DOIs
Publication statusPublished - 1 Jan 2018
MoE publication typeA4 Article in a conference publication
EventIFAC Symposium on System Identification - Stockholm, Sweden
Duration: 9 Jul 201811 Jul 2018
Conference number: 18

Publication series

NameIFAC-PapersOnLine
PublisherElsevier Science Ltd (Pergamon)
ISSN (Print)2405-8963

Conference

ConferenceIFAC Symposium on System Identification
Abbreviated titleSYSID
Country/TerritorySweden
CityStockholm
Period09/07/201811/07/2018

Keywords

  • Bayesian methods
  • estimation
  • filtering
  • Gaussian processes
  • Nonlinear system identification
  • nonparametric methods
  • smoothing

Fingerprint

Dive into the research topics of 'Modeling the Drift Function in Stochastic Differential Equations using Reduced Rank Gaussian Processes'. Together they form a unique fingerprint.

Cite this