斑马:当时间图神经网络满足时间个性化PageRank

Yiming Li, Yanyan Shen, Lei Chen, Mingxuan Yuan
{"title":"斑马:当时间图神经网络满足时间个性化PageRank","authors":"Yiming Li, Yanyan Shen, Lei Chen, Mingxuan Yuan","doi":"10.14778/3583140.3583150","DOIUrl":null,"url":null,"abstract":"\n Temporal graph neural networks (T-GNNs) are state-of-the-art methods for learning representations over dynamic graphs. Despite the superior performance, T-GNNs still suffer from high computational complexity caused by the tedious recursive temporal message passing scheme, which hinders their applicability to large dynamic graphs. To address the problem, we build the theoretical connection between the temporal message passing scheme adopted by T-GNNs and the temporal random walk process on dynamic graphs. Our theoretical analysis indicates that it would be possible to select a few influential temporal neighbors to compute a target node's representation without compromising the predictive performance. Based on this finding, we propose to utilize T-PPR, a parameterized metric for estimating the influence score of nodes on evolving graphs. We further develop an efficient single-scan algorithm to answer the top-\n k\n T-PPR query with rigorous approximation guarantees. Finally, we present Zebra, a scalable framework that accelerates the computation of T-GNN by directly aggregating the features of the most prominent temporal neighbors returned by the top-\n k\n T-PPR query. Extensive experiments have validated that Zebra can be up to two orders of magnitude faster than the state-of-the-art T-GNNs while attaining better performance.\n","PeriodicalId":20467,"journal":{"name":"Proc. VLDB Endow.","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Zebra: When Temporal Graph Neural Networks Meet Temporal Personalized PageRank\",\"authors\":\"Yiming Li, Yanyan Shen, Lei Chen, Mingxuan Yuan\",\"doi\":\"10.14778/3583140.3583150\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"\\n Temporal graph neural networks (T-GNNs) are state-of-the-art methods for learning representations over dynamic graphs. Despite the superior performance, T-GNNs still suffer from high computational complexity caused by the tedious recursive temporal message passing scheme, which hinders their applicability to large dynamic graphs. To address the problem, we build the theoretical connection between the temporal message passing scheme adopted by T-GNNs and the temporal random walk process on dynamic graphs. Our theoretical analysis indicates that it would be possible to select a few influential temporal neighbors to compute a target node's representation without compromising the predictive performance. Based on this finding, we propose to utilize T-PPR, a parameterized metric for estimating the influence score of nodes on evolving graphs. We further develop an efficient single-scan algorithm to answer the top-\\n k\\n T-PPR query with rigorous approximation guarantees. Finally, we present Zebra, a scalable framework that accelerates the computation of T-GNN by directly aggregating the features of the most prominent temporal neighbors returned by the top-\\n k\\n T-PPR query. Extensive experiments have validated that Zebra can be up to two orders of magnitude faster than the state-of-the-art T-GNNs while attaining better performance.\\n\",\"PeriodicalId\":20467,\"journal\":{\"name\":\"Proc. VLDB Endow.\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proc. VLDB Endow.\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.14778/3583140.3583150\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proc. VLDB Endow.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.14778/3583140.3583150","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

摘要

时态图神经网络(t - gnn)是学习动态图表示的最新方法。尽管具有优异的性能,但t - gnn仍然存在繁琐的递归时间消息传递方案导致的高计算复杂度,这阻碍了其在大型动态图中的应用。为了解决这个问题,我们建立了t - gnn采用的时间消息传递方案与动态图上的时间随机漫步过程之间的理论联系。我们的理论分析表明,在不影响预测性能的情况下,选择几个有影响的时间邻居来计算目标节点的表示是可能的。基于这一发现,我们提出利用参数化度量T-PPR来估计节点对进化图的影响评分。我们进一步开发了一种高效的单扫描算法来回答具有严格近似保证的top- k T-PPR查询。最后,我们提出了Zebra,这是一个可扩展的框架,通过直接聚合top- k T-PPR查询返回的最突出的时间邻居的特征来加速T-GNN的计算。大量的实验已经证实,Zebra可以比最先进的t - gnn快两个数量级,同时获得更好的性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Zebra: When Temporal Graph Neural Networks Meet Temporal Personalized PageRank
Temporal graph neural networks (T-GNNs) are state-of-the-art methods for learning representations over dynamic graphs. Despite the superior performance, T-GNNs still suffer from high computational complexity caused by the tedious recursive temporal message passing scheme, which hinders their applicability to large dynamic graphs. To address the problem, we build the theoretical connection between the temporal message passing scheme adopted by T-GNNs and the temporal random walk process on dynamic graphs. Our theoretical analysis indicates that it would be possible to select a few influential temporal neighbors to compute a target node's representation without compromising the predictive performance. Based on this finding, we propose to utilize T-PPR, a parameterized metric for estimating the influence score of nodes on evolving graphs. We further develop an efficient single-scan algorithm to answer the top- k T-PPR query with rigorous approximation guarantees. Finally, we present Zebra, a scalable framework that accelerates the computation of T-GNN by directly aggregating the features of the most prominent temporal neighbors returned by the top- k T-PPR query. Extensive experiments have validated that Zebra can be up to two orders of magnitude faster than the state-of-the-art T-GNNs while attaining better performance.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信