{"about":{"site":"https://codewithpapers.app","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.","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"},"url":"/paper/a-proposal-of-multi-layer-perceptron-with","title":"A Proposal of Multi-Layer Perceptron with Graph Gating Unit for Graph Representation Learning and its Application to Surrogate Model for FEM","arxiv_id":null,"date":"2022-07-11","proceeding":"JSAI 2022 7","authors":["Nakai Yu","Okuda Hiroshi"],"abstract":"GNNs are the neural networks for the representation learning of graph-structured data, most of which are\r\nconstructed by stacking graph convolutional layers. As stacking n-layers of ones is equivalent to propagating n-hop\r\nof neighbor nodes' information, GNNs require enough large number of layers to learn large graphs. However, it\r\ntends to degrade the model performance due to the problem called over-smoothing. In this paper, by presenting\r\na novel GNN model, based on stacking feedforward neural networks with gating structures using GCNs, I tried\r\nto solve the over-smoothing problem and thereby overcome the difficulty of GNNs learning large graphs. The\r\nexperimental results showed that the proposed method monotonically improved the prediction accuracy up to 20\r\nlayers without over-smoothing, whereas the conventional method caused it at 4 to 8 layers. In two experiments on\r\nlarge graphs, the PPI dataset, a benchmark for inductive node classification, and the application to the surrogate\r\nmodel for finite element methods, the proposed method achieved the highest accuracy of the existing methods\r\ncompared, especially with a state-of-the-art accuracy of 99.71% on the PPI dataset.","url_abs":"https://www.jstage.jst.go.jp/article/pjsai/JSAI2022/0/JSAI2022_1G4OS22a03/_article/-char/ja/","url_pdf":"https://www.jstage.jst.go.jp/article/pjsai/JSAI2022/0/JSAI2022_1G4OS22a03/_pdf/-char/ja","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":"a-proposal-of-multi-layer-perceptron-with","repo_url":"https://github.com/nnaakkaaii/g2-MLP","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"graph-representation-learning","task_name":"Graph Representation Learning"},{"task_slug":"node-classification","task_name":"Node Classification"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"spatial-gating-unit","method_name":"Spatial Gating Unit"},{"method_slug":"gmlp","method_name":"gMLP"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/node-classification-on-ppi","task":"Node Classification","dataset":"PPI","model":"g2-MLP","rank_in_archive_order":1,"of":24,"metrics":{"F1":"99.71"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}