FedGT: Federated Node Classification with Scalable Graph Transformer

Graphs are widely used to model relational data. As graphs are getting larger and larger in real-world scenarios, there is a trend to store and compute subgraphs in multiple local systems. For example, recently proposed \emph{subgraph federated learning} methods train Graph Neural Networks (GNNs) di...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Zaixi, Hu, Qingyong, Yu, Yang, Gao, Weibo, Liu, Qi
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 Zhang, Zaixi
Hu, Qingyong
Yu, Yang
Gao, Weibo
Liu, Qi
description Graphs are widely used to model relational data. As graphs are getting larger and larger in real-world scenarios, there is a trend to store and compute subgraphs in multiple local systems. For example, recently proposed \emph{subgraph federated learning} methods train Graph Neural Networks (GNNs) distributively on local subgraphs and aggregate GNN parameters with a central server. However, existing methods have the following limitations: (1) The links between local subgraphs are missing in subgraph federated learning. This could severely damage the performance of GNNs that follow message-passing paradigms to update node/edge features. (2) Most existing methods overlook the subgraph heterogeneity issue, brought by subgraphs being from different parts of the whole graph. To address the aforementioned challenges, we propose a scalable \textbf{Fed}erated \textbf{G}raph \textbf{T}ransformer (\textbf{FedGT}) in the paper. Firstly, we design a hybrid attention scheme to reduce the complexity of the Graph Transformer to linear while ensuring a global receptive field with theoretical bounds. Specifically, each node attends to the sampled local neighbors and a set of curated global nodes to learn both local and global information and be robust to missing links. The global nodes are dynamically updated during training with an online clustering algorithm to capture the data distribution of the corresponding local subgraph. Secondly, FedGT computes clients' similarity based on the aligned global nodes with optimal transport. The similarity is then used to perform weighted averaging for personalized aggregation, which well addresses the data heterogeneity problem. Moreover, local differential privacy is applied to further protect the privacy of clients. Finally, extensive experimental results on 6 datasets and 2 subgraph settings demonstrate the superiority of FedGT.
doi_str_mv 10.48550/arxiv.2401.15203
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2401_15203</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2401_15203</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-c1e977a1ca62c02c336827fc84b421495fff194329bf4079fbf7d23c08fda0d43</originalsourceid><addsrcrecordid>eNotz71uwjAUQGEvDBX0ATrhF0jwX-KYDQUIlRAMZI9ubF9hKRDkRNC-fVvodLYjfYR8cJaqIsvYAuJXuKdCMZ7yTDD5RtZb76p6SX_jI4ze0UPvPC07GIaAwcIY-it9hPFMTxY6aDtPqwi3M60jXAfs48XHGZkgdIN__--U1NtNXe6S_bH6LFf7BHItE8u90Rq4hVxYJqyUeSE02kK1SnBlMkTkRklhWlRMG2xROyEtK9ABc0pOyfy1fTKaWwwXiN_NH6d5cuQPx8JEhw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>FedGT: Federated Node Classification with Scalable Graph Transformer</title><source>arXiv.org</source><creator>Zhang, Zaixi ; Hu, Qingyong ; Yu, Yang ; Gao, Weibo ; Liu, Qi</creator><creatorcontrib>Zhang, Zaixi ; Hu, Qingyong ; Yu, Yang ; Gao, Weibo ; Liu, Qi</creatorcontrib><description>Graphs are widely used to model relational data. As graphs are getting larger and larger in real-world scenarios, there is a trend to store and compute subgraphs in multiple local systems. For example, recently proposed \emph{subgraph federated learning} methods train Graph Neural Networks (GNNs) distributively on local subgraphs and aggregate GNN parameters with a central server. However, existing methods have the following limitations: (1) The links between local subgraphs are missing in subgraph federated learning. This could severely damage the performance of GNNs that follow message-passing paradigms to update node/edge features. (2) Most existing methods overlook the subgraph heterogeneity issue, brought by subgraphs being from different parts of the whole graph. To address the aforementioned challenges, we propose a scalable \textbf{Fed}erated \textbf{G}raph \textbf{T}ransformer (\textbf{FedGT}) in the paper. Firstly, we design a hybrid attention scheme to reduce the complexity of the Graph Transformer to linear while ensuring a global receptive field with theoretical bounds. Specifically, each node attends to the sampled local neighbors and a set of curated global nodes to learn both local and global information and be robust to missing links. The global nodes are dynamically updated during training with an online clustering algorithm to capture the data distribution of the corresponding local subgraph. Secondly, FedGT computes clients' similarity based on the aligned global nodes with optimal transport. The similarity is then used to perform weighted averaging for personalized aggregation, which well addresses the data heterogeneity problem. Moreover, local differential privacy is applied to further protect the privacy of clients. Finally, extensive experimental results on 6 datasets and 2 subgraph settings demonstrate the superiority of FedGT.</description><identifier>DOI: 10.48550/arxiv.2401.15203</identifier><language>eng</language><subject>Computer Science - Learning</subject><creationdate>2024-01</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2401.15203$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2401.15203$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhang, Zaixi</creatorcontrib><creatorcontrib>Hu, Qingyong</creatorcontrib><creatorcontrib>Yu, Yang</creatorcontrib><creatorcontrib>Gao, Weibo</creatorcontrib><creatorcontrib>Liu, Qi</creatorcontrib><title>FedGT: Federated Node Classification with Scalable Graph Transformer</title><description>Graphs are widely used to model relational data. As graphs are getting larger and larger in real-world scenarios, there is a trend to store and compute subgraphs in multiple local systems. For example, recently proposed \emph{subgraph federated learning} methods train Graph Neural Networks (GNNs) distributively on local subgraphs and aggregate GNN parameters with a central server. However, existing methods have the following limitations: (1) The links between local subgraphs are missing in subgraph federated learning. This could severely damage the performance of GNNs that follow message-passing paradigms to update node/edge features. (2) Most existing methods overlook the subgraph heterogeneity issue, brought by subgraphs being from different parts of the whole graph. To address the aforementioned challenges, we propose a scalable \textbf{Fed}erated \textbf{G}raph \textbf{T}ransformer (\textbf{FedGT}) in the paper. Firstly, we design a hybrid attention scheme to reduce the complexity of the Graph Transformer to linear while ensuring a global receptive field with theoretical bounds. Specifically, each node attends to the sampled local neighbors and a set of curated global nodes to learn both local and global information and be robust to missing links. The global nodes are dynamically updated during training with an online clustering algorithm to capture the data distribution of the corresponding local subgraph. Secondly, FedGT computes clients' similarity based on the aligned global nodes with optimal transport. The similarity is then used to perform weighted averaging for personalized aggregation, which well addresses the data heterogeneity problem. Moreover, local differential privacy is applied to further protect the privacy of clients. Finally, extensive experimental results on 6 datasets and 2 subgraph settings demonstrate the superiority of FedGT.</description><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz71uwjAUQGEvDBX0ATrhF0jwX-KYDQUIlRAMZI9ubF9hKRDkRNC-fVvodLYjfYR8cJaqIsvYAuJXuKdCMZ7yTDD5RtZb76p6SX_jI4ze0UPvPC07GIaAwcIY-it9hPFMTxY6aDtPqwi3M60jXAfs48XHGZkgdIN__--U1NtNXe6S_bH6LFf7BHItE8u90Rq4hVxYJqyUeSE02kK1SnBlMkTkRklhWlRMG2xROyEtK9ABc0pOyfy1fTKaWwwXiN_NH6d5cuQPx8JEhw</recordid><startdate>20240126</startdate><enddate>20240126</enddate><creator>Zhang, Zaixi</creator><creator>Hu, Qingyong</creator><creator>Yu, Yang</creator><creator>Gao, Weibo</creator><creator>Liu, Qi</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240126</creationdate><title>FedGT: Federated Node Classification with Scalable Graph Transformer</title><author>Zhang, Zaixi ; Hu, Qingyong ; Yu, Yang ; Gao, Weibo ; Liu, Qi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-c1e977a1ca62c02c336827fc84b421495fff194329bf4079fbf7d23c08fda0d43</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Zaixi</creatorcontrib><creatorcontrib>Hu, Qingyong</creatorcontrib><creatorcontrib>Yu, Yang</creatorcontrib><creatorcontrib>Gao, Weibo</creatorcontrib><creatorcontrib>Liu, Qi</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhang, Zaixi</au><au>Hu, Qingyong</au><au>Yu, Yang</au><au>Gao, Weibo</au><au>Liu, Qi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>FedGT: Federated Node Classification with Scalable Graph Transformer</atitle><date>2024-01-26</date><risdate>2024</risdate><abstract>Graphs are widely used to model relational data. As graphs are getting larger and larger in real-world scenarios, there is a trend to store and compute subgraphs in multiple local systems. For example, recently proposed \emph{subgraph federated learning} methods train Graph Neural Networks (GNNs) distributively on local subgraphs and aggregate GNN parameters with a central server. However, existing methods have the following limitations: (1) The links between local subgraphs are missing in subgraph federated learning. This could severely damage the performance of GNNs that follow message-passing paradigms to update node/edge features. (2) Most existing methods overlook the subgraph heterogeneity issue, brought by subgraphs being from different parts of the whole graph. To address the aforementioned challenges, we propose a scalable \textbf{Fed}erated \textbf{G}raph \textbf{T}ransformer (\textbf{FedGT}) in the paper. Firstly, we design a hybrid attention scheme to reduce the complexity of the Graph Transformer to linear while ensuring a global receptive field with theoretical bounds. Specifically, each node attends to the sampled local neighbors and a set of curated global nodes to learn both local and global information and be robust to missing links. The global nodes are dynamically updated during training with an online clustering algorithm to capture the data distribution of the corresponding local subgraph. Secondly, FedGT computes clients' similarity based on the aligned global nodes with optimal transport. The similarity is then used to perform weighted averaging for personalized aggregation, which well addresses the data heterogeneity problem. Moreover, local differential privacy is applied to further protect the privacy of clients. Finally, extensive experimental results on 6 datasets and 2 subgraph settings demonstrate the superiority of FedGT.</abstract><doi>10.48550/arxiv.2401.15203</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2401.15203
ispartof
issn
language eng
recordid cdi_arxiv_primary_2401_15203
source arXiv.org
subjects Computer Science - Learning
title FedGT: Federated Node Classification with Scalable Graph Transformer
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T20%3A43%3A50IST&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=FedGT:%20Federated%20Node%20Classification%20with%20Scalable%20Graph%20Transformer&rft.au=Zhang,%20Zaixi&rft.date=2024-01-26&rft_id=info:doi/10.48550/arxiv.2401.15203&rft_dat=%3Carxiv_GOX%3E2401_15203%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