Attentional Gated Res2Net for Multivariate Time Series Classification

Yang, C; Wang, X; Yao, L; Long, G; Jiang, J; Xu, G

Attentional Gated Res2Net for Multivariate Time Series Classification

Yang, C Wang, X

Yao, L Long, G Jiang, J Xu, G

Permalink

Publisher:: Springer
Publication Type:: Journal Article
Citation:: Neural Processing Letters, 2022, pp. 1-25
Issue Date:: 2022-12-30

Recently Added

	Filename	Description	Size
	AGRES2NET_Neural_Processing_Letters.pdf	Accepted version	493.03 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is new to OPUS and is not currently available.

Full metadata record

Field	Value	Language
dc.contributor.author	Yang, C
dc.contributor.author	Wang, X https://orcid.org/0000-0001-9582-3445
dc.contributor.author	Yao, L
dc.contributor.author	Long, G
dc.contributor.author	Jiang, J
dc.contributor.author	Xu, G
dc.date.accessioned	2022-07-08T06:19:10Z
dc.date.available	2022-07-08T06:19:10Z
dc.date.issued	2022-12-30
dc.identifier.citation	Neural Processing Letters, 2022, pp. 1-25
dc.identifier.issn	1370-4621
dc.identifier.issn	1573-773X
dc.identifier.uri	http://hdl.handle.net/10453/158759
dc.description.abstract	<jats:title>Abstract</jats:title><jats:p>Multivariate time series classification is a critical problem in data mining with broad applications. It requires harnessing the inter-relationship of multiple variables and various ranges of temporal dependencies to assign the correct classification label of the time series. Multivariate time series may come from a wide range of sources and be used in various scenarios, bringing the classifier challenge of temporal representation learning. We propose a novel convolutional neural network architecture called Attentional Gated Res2Net for multivariate time series classification. Our model uses hierarchical residual-like connections to achieve multi-scale receptive fields and capture multi-granular temporal information. The gating mechanism enables the model to consider the relations between the feature maps extracted by receptive fields of multiple sizes for information fusion. Further, we propose two types of attention modules, channel-wise attention and block-wise attention, to better leverage the multi-granular temporal patterns. Our experimental results on 14 benchmark multivariate time-series datasets show that our model outperforms several baselines and state-of-the-art methods by a large margin. Our model outperforms the SOTA by a large margin, the classification accuracy of our model is 10.16% better than the SOTA model. Besides, we demonstrate that our model improves the performance of existing models when used as a plugin. Further, based on our experiments and analysis, we provide practical advice on applying our model to a new problem.</jats:p>
dc.language	en
dc.publisher	Springer
dc.relation	http://purl.org/au-research/grants/arc/LP180100654
dc.relation	http://purl.org/au-research/grants/arc/DE180100251
dc.relation	http://purl.org/au-research/grants/arc/LE220100078
dc.relation	http://purl.org/au-research/grants/arc/DP220103717
dc.relation.ispartof	Neural Processing Letters
dc.relation.isbasedon	10.1007/s11063-022-10944-0
dc.rights	info:eu-repo/semantics/restrictedAccess
dc.rights	Springer This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://link.springer.com/article/10.1007/s11063-022-10944-0
dc.subject	0801 Artificial Intelligence and Image Processing, 1702 Cognitive Sciences
dc.subject.classification	Artificial Intelligence & Image Processing
dc.title	Attentional Gated Res2Net for Multivariate Time Series Classification
dc.type	Journal Article
utslib.for	0801 Artificial Intelligence and Image Processing
utslib.for	1702 Cognitive Sciences
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	recently_added	*
pubs.consider-herdc	false
dc.date.updated	2022-07-08T06:19:08Z
pubs.publication-status	Accepted

Abstract:

AbstractMultivariate time series classification is a critical problem in data mining with broad applications. It requires harnessing the inter-relationship of multiple variables and various ranges of temporal dependencies to assign the correct classification label of the time series. Multivariate time series may come from a wide range of sources and be used in various scenarios, bringing the classifier challenge of temporal representation learning. We propose a novel convolutional neural network architecture called Attentional Gated Res2Net for multivariate time series classification. Our model uses hierarchical residual-like connections to achieve multi-scale receptive fields and capture multi-granular temporal information. The gating mechanism enables the model to consider the relations between the feature maps extracted by receptive fields of multiple sizes for information fusion. Further, we propose two types of attention modules, channel-wise attention and block-wise attention, to better leverage the multi-granular temporal patterns. Our experimental results on 14 benchmark multivariate time-series datasets show that our model outperforms several baselines and state-of-the-art methods by a large margin. Our model outperforms the SOTA by a large margin, the classification accuracy of our model is 10.16% better than the SOTA model. Besides, we demonstrate that our model improves the performance of existing models when used as a plugin. Further, based on our experiments and analysis, we provide practical advice on applying our model to a new problem.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/158759