{"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 machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/the-split-matters-flat-minima-methods-for","title":"The Split Matters: Flat Minima Methods for Improving the Performance of GNNs","arxiv_id":"2306.09121","date":"2023-06-15","proceeding":null,"authors":["Nicolas Lell","Ansgar Scherp"],"abstract":"When training a Neural Network, it is optimized using the available training data with the hope that it generalizes well to new or unseen testing data. At the same absolute value, a flat minimum in the loss landscape is presumed to generalize better than a sharp minimum. Methods for determining flat minima have been mostly researched for independent and identically distributed (i. i. d.) data such as images. Graphs are inherently non-i. i. d. since the vertices are edge-connected. We investigate flat minima methods and combinations of those methods for training graph neural networks (GNNs). We use GCN and GAT as well as extend Graph-MLP to work with more layers and larger graphs. We conduct experiments on small and large citation, co-purchase, and protein datasets with different train-test splits in both the transductive and inductive training procedure. Results show that flat minima methods can improve the performance of GNN models by over 2 points, if the train-test split is randomized. Following Shchur et al., randomized splits are essential for a fair evaluation of GNNs, as other (fixed) splits like 'Planetoid' are biased. Overall, we provide important insights for improving and fairly evaluating flat minima methods on GNNs. We recommend practitioners to always use weight averaging techniques, in particular EWA when using early stopping. While weight averaging techniques are only sometimes the best performing method, they are less sensitive to hyperparameters, need no additional training, and keep the original model unchanged. All source code is available in https://github.com/Foisunt/FMMs-in-GNNs.","url_abs":"https://arxiv.org/abs/2306.09121v1","url_pdf":"https://arxiv.org/pdf/2306.09121v1.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":"the-split-matters-flat-minima-methods-for","repo_url":"https://github.com/foisunt/fmms-in-gnns","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"node-classification","task_name":"Node Classification"}],"methods":[{"method_slug":"gat","method_name":"GAT"},{"method_slug":"gcn","method_name":"GCN"},{"method_slug":"sharpness-aware-minimization","method_name":"Sharpness-Aware Minimization"},{"method_slug":"stochastic-weight-averaging","method_name":"Stochastic Weight Averaging"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/node-classification-on-citeseer-with-public","task":"Node Classification","dataset":"CiteSeer with Public Split: fixed 20 nodes per class","model":"Graph-MLP + PGN","rank_in_archive_order":4,"of":40,"metrics":{"Accuracy":"74.73 ± 0.6%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-citeseer","task":"Node Classification","dataset":"Citeseer","model":"Graph-MLP + SWA","rank_in_archive_order":9,"of":71,"metrics":{"Accuracy":"77.99 ± 1.57%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora","task":"Node Classification","dataset":"Cora","model":"GAT + SWA","rank_in_archive_order":7,"of":73,"metrics":{"Accuracy":"88.66 ± 1.38%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-cora-with-public-split","task":"Node Classification","dataset":"Cora with Public Split: fixed 20 nodes per class","model":"GAT+PGN","rank_in_archive_order":19,"of":36,"metrics":{"Accuracy":"83.26 ± 0.69%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-ppi","task":"Node Classification","dataset":"PPI","model":"GCN + SAF","rank_in_archive_order":8,"of":24,"metrics":{"F1":"99.38 ± 0.01%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-ppi","task":"Node Classification","dataset":"PPI","model":"GAT + PGN","rank_in_archive_order":10,"of":24,"metrics":{"F1":"99.34 ± 0.02%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed-60-20-20-random","task":"Node Classification","dataset":"PubMed (60%/20%/20% random splits)","model":"Graph-MLP + SAF","rank_in_archive_order":9,"of":37,"metrics":{"1:1 Accuracy":"90.64 ± 0.46%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed-with-public","task":"Node Classification","dataset":"PubMed with Public Split: fixed 20 nodes per class","model":"Graph-MLP + ASAM","rank_in_archive_order":4,"of":37,"metrics":{"Accuracy":"82.60 ± 0.80%"},"uses_additional_data":false},{"leaderboard":"/sota/node-classification-on-pubmed","task":"Node Classification","dataset":"Pubmed","model":"Graph-MLP + SAF","rank_in_archive_order":5,"of":70,"metrics":{"Accuracy":"90.64 ± 0.46%"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers","held":null,"reason":"arxiv_id_not_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"by_repo":[{"repo":"foisunt/fmms-in-gnns","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"ddb03334ab9e9188b55dac5569e3d1e10e1b312848fa57d6ad51626f0101bfc9","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}