A Nonparametric Spatio-temporal SDE Model

Cagatay Yildiz; Markus Heinonen; Harri Lähdesmäki

A Nonparametric Spatio-temporal SDE Model

Cagatay Yildiz, Markus Heinonen, Harri Lähdesmäki

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference article in proceedings › Professional

Abstrakti

We propose a nonparametric spatio-temporal stochastic differential equation (SDE) model that can learn the underlying dynamics of arbitrary continuous-time systems without prior knowledge. We augment the input space of the drift function of an SDE with a temporal component to account for spatio-temporal patterns. The experiments on a real world data set demonstrate that the spatio-temporal model is better able to fit the data than the spatial model and also reduce the forecasting error.

Alkuperäiskieli	Englanti
Otsikko	NIPS 2018 Spatiotemporal Workshop
Alaotsikko	32nd Conference on Neural Information Processing Systems (NIPS 2018), Montréal, Canada
Kustantaja	Neural Information Processing Systems Foundation
Sivut	1-5
Tila	Julkaistu - 2018
OKM-julkaisutyyppi	D3 Artikkeli ammatillisessa konferenssijulkaisussa
Tapahtuma	NIPS Spatiotemporal Workshop - Montreal, Kanada Kesto: 3 jouluk. 2018 → 8 jouluk. 2018

Workshop

Workshop	NIPS Spatiotemporal Workshop
Maa/Alue	Kanada
Kaupunki	Montreal
Ajanjakso	03/12/2018 → 08/12/2018

Muut tiedostot ja linkit

https://openreview.net/forum?id=HJzoRCKwjQ

Siteeraa tätä

@inproceedings{75a5e67643bf410daa0cb88cfb107687,

title = "A Nonparametric Spatio-temporal SDE Model",

abstract = "We propose a nonparametric spatio-temporal stochastic differential equation (SDE) model that can learn the underlying dynamics of arbitrary continuous-time systems without prior knowledge. We augment the input space of the drift function of an SDE with a temporal component to account for spatio-temporal patterns. The experiments on a real world data set demonstrate that the spatio-temporal model is better able to fit the data than the spatial model and also reduce the forecasting error.",

keywords = "stochastic differential equation, gaussian processes, spatiotemporal drift",

author = "Cagatay Yildiz and Markus Heinonen and Harri L{\"a}hdesm{\"a}ki",

year = "2018",

language = "English",

pages = "1--5",

booktitle = "NIPS 2018 Spatiotemporal Workshop",

publisher = "Neural Information Processing Systems Foundation",

address = "United States",

note = "NIPS Spatiotemporal Workshop ; Conference date: 03-12-2018 Through 08-12-2018",

}

A Nonparametric Spatio-temporal SDE Model. / Yildiz, Cagatay; Heinonen, Markus ; Lähdesmäki, Harri.
NIPS 2018 Spatiotemporal Workshop: 32nd Conference on Neural Information Processing Systems (NIPS 2018), Montréal, Canada. Neural Information Processing Systems Foundation, 2018. s. 1-5.

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference article in proceedings › Professional

TY - GEN

T1 - A Nonparametric Spatio-temporal SDE Model

AU - Yildiz, Cagatay

AU - Heinonen, Markus

AU - Lähdesmäki, Harri

PY - 2018

Y1 - 2018

N2 - We propose a nonparametric spatio-temporal stochastic differential equation (SDE) model that can learn the underlying dynamics of arbitrary continuous-time systems without prior knowledge. We augment the input space of the drift function of an SDE with a temporal component to account for spatio-temporal patterns. The experiments on a real world data set demonstrate that the spatio-temporal model is better able to fit the data than the spatial model and also reduce the forecasting error.

AB - We propose a nonparametric spatio-temporal stochastic differential equation (SDE) model that can learn the underlying dynamics of arbitrary continuous-time systems without prior knowledge. We augment the input space of the drift function of an SDE with a temporal component to account for spatio-temporal patterns. The experiments on a real world data set demonstrate that the spatio-temporal model is better able to fit the data than the spatial model and also reduce the forecasting error.

KW - stochastic differential equation

KW - gaussian processes

KW - spatiotemporal drift

UR - https://openreview.net/forum?id=HJzoRCKwjQ

M3 - Conference article in proceedings

SP - 1

EP - 5

BT - NIPS 2018 Spatiotemporal Workshop

PB - Neural Information Processing Systems Foundation

T2 - NIPS Spatiotemporal Workshop

Y2 - 3 December 2018 through 8 December 2018

ER -

A Nonparametric Spatio-temporal SDE Model

Abstrakti

Workshop

Muut tiedostot ja linkit

Sormenjälki

Siteeraa tätä