Show simple item record

dc.contributor.author
Cao, Defu
dc.contributor.author
Wang, Yujing
dc.contributor.author
Duan, Juanyong
dc.contributor.author
Zhang, Ce
dc.contributor.author
Zhu, Xia
dc.contributor.author
Huang, Congrui
dc.contributor.author
Tong, Yunhai
dc.contributor.author
Xu, Bixiong
dc.contributor.author
Bai, Jing
dc.contributor.author
Tong, Jie
dc.contributor.author
Zhang, Qi
dc.contributor.editor
Larochelle, Hugo
dc.contributor.editor
Ranzato, Marc'Aurelio
dc.contributor.editor
Hadsell, Raia
dc.contributor.editor
Balcan, Maria F.
dc.contributor.editor
Lin, H.
dc.date.accessioned
2021-07-21T09:21:11Z
dc.date.available
2021-01-22T12:47:29Z
dc.date.available
2021-02-12T13:09:31Z
dc.date.available
2021-03-02T15:27:00Z
dc.date.available
2021-07-21T09:21:11Z
dc.date.issued
2021
dc.identifier.isbn
978-1-7138-2954-6
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/464779
dc.description.abstract
Multivariate time-series forecasting plays a crucial role in many real-world applications. It is a challenging problem as one needs to consider both intra-series temporal correlations and inter-series correlations simultaneously. Recently, there have been multiple works trying to capture both correlations, but most, if not all of them only capture temporal correlations in the time domain and resort to pre-defined priors as inter-series relationships.</p> <p>In this paper, we propose Spectral Temporal Graph Neural Network (StemGNN) to further improve the accuracy of multivariate time-series forecasting. StemGNN captures inter-series correlations and temporal dependencies jointly in the spectral domain. It combines Graph Fourier Transform (GFT) which models inter-series correlations and Discrete Fourier Transform (DFT) which models temporal dependencies in an end-to-end framework. After passing through GFT and DFT, the spectral representations hold clear patterns and can be predicted effectively by convolution and sequential learning modules. Moreover, StemGNN learns inter-series correlations automatically from the data without using pre-defined priors. We conduct extensive experiments on ten real-world datasets to demonstrate the effectiveness of StemGNN.
en_US
dc.language.iso
en
en_US
dc.publisher
Curran
en_US
dc.title
Spectral Temporal Graph Neural Network for Multivariate Time-series Forecasting
en_US
dc.type
Conference Paper
dc.date.published
2020
ethz.book.title
Advances in Neural Information Processing Systems 33
en_US
ethz.pages.start
17766
en_US
ethz.pages.end
17778
en_US
ethz.event
34th Annual Conference on Neural Information Processing Systems (NeurIPS 2020)
en_US
ethz.event.location
Online
en_US
ethz.event.date
December 6-12, 2020
en_US
ethz.notes
Due to the Coronavirus (COVID-19) the conference was conducted virtually.
en_US
ethz.publication.place
Red Hook, NY
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02150 - Dep. Informatik / Dep. of Computer Science::02663 - Institut für Computing Platforms / Institute for Computing Platforms::09588 - Zhang, Ce / Zhang, Ce
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02150 - Dep. Informatik / Dep. of Computer Science::02663 - Institut für Computing Platforms / Institute for Computing Platforms::09588 - Zhang, Ce / Zhang, Ce
en_US
ethz.identifier.url
https://papers.nips.cc/paper/2020/hash/cdf6581cb7aca4b7e19ef136c6e601a5-Abstract.html
ethz.date.deposited
2021-01-22T12:47:36Z
ethz.source
FORM
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2021-03-02T15:27:10Z
ethz.rosetta.lastUpdated
2022-03-29T10:34:27Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&amp;rft_val_fmt=info:ofi/fmt:kev:mtx:journal&amp;rft.atitle=Spectral%20Temporal%20Graph%20Neural%20Network%20for%20Multivariate%20Time-series%20Forecasting&amp;rft.date=2021&amp;rft.spage=17766&amp;rft.epage=17778&amp;rft.au=Cao,%20Defu&amp;Wang,%20Yujing&amp;Duan,%20Juanyong&amp;Zhang,%20Ce&amp;Zhu,%20Xia&amp;rft.isbn=978-1-7138-2954-6&amp;rft.genre=proceeding&amp;rft.btitle=Advances%20in%20Neural%20Information%20Processing%20Systems%2033
 Search print copy at ETH Library

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record