Towards Generating Real-World Time Series Data
Time series data generation has drawn increasing attention in recent years. Several generative adversarial network (GAN) based methods have been proposed to tackle the problem usually with the assumption that the targeted time series data are well-formatted and complete. However, real-world time ser...
Gespeichert in:
Hauptverfasser: | , , , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | |
container_volume | |
creator | Pei, Hengzhi Ren, Kan Yang, Yuqing Liu, Chang Qin, Tao Li, Dongsheng |
description | Time series data generation has drawn increasing attention in recent years.
Several generative adversarial network (GAN) based methods have been proposed
to tackle the problem usually with the assumption that the targeted time series
data are well-formatted and complete. However, real-world time series (RTS)
data are far away from this utopia, e.g., long sequences with variable lengths
and informative missing data raise intractable challenges for designing
powerful generation algorithms. In this paper, we propose a novel generative
framework for RTS data - RTSGAN to tackle the aforementioned challenges. RTSGAN
first learns an encoder-decoder module which provides a mapping between a time
series instance and a fixed-dimension latent vector and then learns a
generation module to generate vectors in the same latent space. By combining
the generator and the decoder, RTSGAN is able to generate RTS which respect the
original feature distributions and the temporal dynamics. To generate time
series with missing values, we further equip RTSGAN with an observation
embedding layer and a decide-and-generate decoder to better utilize the
informative missing patterns. Experiments on the four RTS datasets show that
the proposed framework outperforms the previous generation methods in terms of
synthetic data utility for downstream classification and prediction tasks. |
doi_str_mv | 10.48550/arxiv.2111.08386 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2111_08386</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2111_08386</sourcerecordid><originalsourceid>FETCH-LOGICAL-a676-f156e75807c847eca9f64063b22fbe839e2755c3a7ae6f79b8bbac14016d07683</originalsourceid><addsrcrecordid>eNotzsuKwjAUgOFsXAzqA8zKvEBr0jQn6VLqZQYEQQsuy0l7IoFqJRUvbz-MM6t_9_Mx9ilFmlutxRzjM9zTTEqZCqssfLC06h8Y24Fv6EIRb-Fy4nvCLjn2sWt5Fc7EDxQDDXyJN5ywkcduoOl_x6xar6ryK9nuNt_lYpsgGEi81EBGW2EamxtqsPCQC1Auy7wjqwrKjNaNQoME3hTOOoeNzIWEVhiwasxmf9s3uL7GcMb4qn_h9RuufgAt1DxP</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Towards Generating Real-World Time Series Data</title><source>arXiv.org</source><creator>Pei, Hengzhi ; Ren, Kan ; Yang, Yuqing ; Liu, Chang ; Qin, Tao ; Li, Dongsheng</creator><creatorcontrib>Pei, Hengzhi ; Ren, Kan ; Yang, Yuqing ; Liu, Chang ; Qin, Tao ; Li, Dongsheng</creatorcontrib><description>Time series data generation has drawn increasing attention in recent years.
Several generative adversarial network (GAN) based methods have been proposed
to tackle the problem usually with the assumption that the targeted time series
data are well-formatted and complete. However, real-world time series (RTS)
data are far away from this utopia, e.g., long sequences with variable lengths
and informative missing data raise intractable challenges for designing
powerful generation algorithms. In this paper, we propose a novel generative
framework for RTS data - RTSGAN to tackle the aforementioned challenges. RTSGAN
first learns an encoder-decoder module which provides a mapping between a time
series instance and a fixed-dimension latent vector and then learns a
generation module to generate vectors in the same latent space. By combining
the generator and the decoder, RTSGAN is able to generate RTS which respect the
original feature distributions and the temporal dynamics. To generate time
series with missing values, we further equip RTSGAN with an observation
embedding layer and a decide-and-generate decoder to better utilize the
informative missing patterns. Experiments on the four RTS datasets show that
the proposed framework outperforms the previous generation methods in terms of
synthetic data utility for downstream classification and prediction tasks.</description><identifier>DOI: 10.48550/arxiv.2111.08386</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning</subject><creationdate>2021-11</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,782,887</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2111.08386$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2111.08386$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Pei, Hengzhi</creatorcontrib><creatorcontrib>Ren, Kan</creatorcontrib><creatorcontrib>Yang, Yuqing</creatorcontrib><creatorcontrib>Liu, Chang</creatorcontrib><creatorcontrib>Qin, Tao</creatorcontrib><creatorcontrib>Li, Dongsheng</creatorcontrib><title>Towards Generating Real-World Time Series Data</title><description>Time series data generation has drawn increasing attention in recent years.
Several generative adversarial network (GAN) based methods have been proposed
to tackle the problem usually with the assumption that the targeted time series
data are well-formatted and complete. However, real-world time series (RTS)
data are far away from this utopia, e.g., long sequences with variable lengths
and informative missing data raise intractable challenges for designing
powerful generation algorithms. In this paper, we propose a novel generative
framework for RTS data - RTSGAN to tackle the aforementioned challenges. RTSGAN
first learns an encoder-decoder module which provides a mapping between a time
series instance and a fixed-dimension latent vector and then learns a
generation module to generate vectors in the same latent space. By combining
the generator and the decoder, RTSGAN is able to generate RTS which respect the
original feature distributions and the temporal dynamics. To generate time
series with missing values, we further equip RTSGAN with an observation
embedding layer and a decide-and-generate decoder to better utilize the
informative missing patterns. Experiments on the four RTS datasets show that
the proposed framework outperforms the previous generation methods in terms of
synthetic data utility for downstream classification and prediction tasks.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzsuKwjAUgOFsXAzqA8zKvEBr0jQn6VLqZQYEQQsuy0l7IoFqJRUvbz-MM6t_9_Mx9ilFmlutxRzjM9zTTEqZCqssfLC06h8Y24Fv6EIRb-Fy4nvCLjn2sWt5Fc7EDxQDDXyJN5ywkcduoOl_x6xar6ryK9nuNt_lYpsgGEi81EBGW2EamxtqsPCQC1Auy7wjqwrKjNaNQoME3hTOOoeNzIWEVhiwasxmf9s3uL7GcMb4qn_h9RuufgAt1DxP</recordid><startdate>20211116</startdate><enddate>20211116</enddate><creator>Pei, Hengzhi</creator><creator>Ren, Kan</creator><creator>Yang, Yuqing</creator><creator>Liu, Chang</creator><creator>Qin, Tao</creator><creator>Li, Dongsheng</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20211116</creationdate><title>Towards Generating Real-World Time Series Data</title><author>Pei, Hengzhi ; Ren, Kan ; Yang, Yuqing ; Liu, Chang ; Qin, Tao ; Li, Dongsheng</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a676-f156e75807c847eca9f64063b22fbe839e2755c3a7ae6f79b8bbac14016d07683</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Pei, Hengzhi</creatorcontrib><creatorcontrib>Ren, Kan</creatorcontrib><creatorcontrib>Yang, Yuqing</creatorcontrib><creatorcontrib>Liu, Chang</creatorcontrib><creatorcontrib>Qin, Tao</creatorcontrib><creatorcontrib>Li, Dongsheng</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Pei, Hengzhi</au><au>Ren, Kan</au><au>Yang, Yuqing</au><au>Liu, Chang</au><au>Qin, Tao</au><au>Li, Dongsheng</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Towards Generating Real-World Time Series Data</atitle><date>2021-11-16</date><risdate>2021</risdate><abstract>Time series data generation has drawn increasing attention in recent years.
Several generative adversarial network (GAN) based methods have been proposed
to tackle the problem usually with the assumption that the targeted time series
data are well-formatted and complete. However, real-world time series (RTS)
data are far away from this utopia, e.g., long sequences with variable lengths
and informative missing data raise intractable challenges for designing
powerful generation algorithms. In this paper, we propose a novel generative
framework for RTS data - RTSGAN to tackle the aforementioned challenges. RTSGAN
first learns an encoder-decoder module which provides a mapping between a time
series instance and a fixed-dimension latent vector and then learns a
generation module to generate vectors in the same latent space. By combining
the generator and the decoder, RTSGAN is able to generate RTS which respect the
original feature distributions and the temporal dynamics. To generate time
series with missing values, we further equip RTSGAN with an observation
embedding layer and a decide-and-generate decoder to better utilize the
informative missing patterns. Experiments on the four RTS datasets show that
the proposed framework outperforms the previous generation methods in terms of
synthetic data utility for downstream classification and prediction tasks.</abstract><doi>10.48550/arxiv.2111.08386</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.2111.08386 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_2111_08386 |
source | arXiv.org |
subjects | Computer Science - Artificial Intelligence Computer Science - Learning |
title | Towards Generating Real-World Time Series Data |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-03T05%3A57%3A29IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Towards%20Generating%20Real-World%20Time%20Series%20Data&rft.au=Pei,%20Hengzhi&rft.date=2021-11-16&rft_id=info:doi/10.48550/arxiv.2111.08386&rft_dat=%3Carxiv_GOX%3E2111_08386%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |