{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the layout of this site's pages, twins and files"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts per paper"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer and the archive sources"}]},"url":"/paper/large-scale-learnable-graph-convolutional","title":"Large-Scale Learnable Graph Convolutional Networks","arxiv_id":"1808.03965","date":"2018-08-12","proceeding":null,"authors":["Hongyang Gao","Zhengyang Wang","Shuiwang Ji"],"abstract":"Convolutional neural networks (CNNs) have achieved great success on grid-like\ndata such as images, but face tremendous challenges in learning from more\ngeneric data such as graphs. In CNNs, the trainable local filters enable the\nautomatic extraction of high-level features. The computation with filters\nrequires a fixed number of ordered units in the receptive fields. However, the\nnumber of neighboring units is neither fixed nor are they ordered in generic\ngraphs, thereby hindering the applications of convolutional operations. Here,\nwe address these challenges by proposing the learnable graph convolutional\nlayer (LGCL). LGCL automatically selects a fixed number of neighboring nodes\nfor each feature based on value ranking in order to transform graph data into\ngrid-like structures in 1-D format, thereby enabling the use of regular\nconvolutional operations on generic graphs. To enable model training on\nlarge-scale graphs, we propose a sub-graph training method to reduce the\nexcessive memory and computational resource requirements suffered by prior\nmethods on graph convolutions. Our experimental results on node classification\ntasks in both transductive and inductive learning settings demonstrate that our\nmethods can achieve consistently better performance on the Cora, Citeseer,\nPubmed citation network, and protein-protein interaction network datasets. Our\nresults also indicate that the proposed methods using sub-graph training\nstrategy are more efficient as compared to prior approaches.","url_abs":"http://arxiv.org/abs/1808.03965v1","url_pdf":"http://arxiv.org/pdf/1808.03965v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"large-scale-learnable-graph-convolutional","repo_url":"https://github.com/divelab/lgcn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"document-classification","task_name":"Document Classification"},{"task_slug":"inductive-learning","task_name":"Inductive Learning"},{"task_slug":"node-classification","task_name":"Node Classification"}],"methods":[{"method_slug":"lgcl","method_name":"LGCL"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/document-classification-on-cora","task":"Document Classification","dataset":"Cora","model":"LGCN","rank_in_archive_order":2,"of":6,"metrics":{"Accuracy":"83.3%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer","task":"Node Classification","dataset":"Citeseer","model":"LGCN sub","rank_in_archive_order":37,"of":71,"metrics":{"Accuracy":"73.0 ± 0.6%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora","task":"Node Classification","dataset":"Cora","model":"LGCN sub","rank_in_archive_order":40,"of":73,"metrics":{"Accuracy":"83.3%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-ppi","task":"Node Classification","dataset":"PPI","model":"LGCN","rank_in_archive_order":19,"of":24,"metrics":{"F1":"77.2"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed","task":"Node Classification","dataset":"Pubmed","model":"LGCN sub","rank_in_archive_order":43,"of":70,"metrics":{"Accuracy":"79.5%"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1808.03965","atlas_url":"https://app.syntology.ai/?focus=1808.03965","mcp":null,"developers":"https://syntology.ai/developers"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null}