{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YX3IW4BWPLODF3ZRRVKWSDRVKV","short_pith_number":"pith:YX3IW4BW","schema_version":"1.0","canonical_sha256":"c5f68b70367adc32ef318d55690e355546d8f26a93a9c44edd6bffe7772b6628","source":{"kind":"arxiv","id":"2306.09121","version":1},"attestation_state":"computed","paper":{"title":"The Split Matters: Flat Minima Methods for Improving the Performance of GNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ansgar Scherp, Nicolas Lell","submitted_at":"2023-06-15T13:29:09Z","abstract_excerpt":"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"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.09121","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-15T13:29:09Z","cross_cats_sorted":[],"title_canon_sha256":"f6c41317119eca279d44e8933b638208591e9bcd1644eb5850553efed293362b","abstract_canon_sha256":"2284f3a8bcc672b0d950e667475f9d3793b7d3b8df1d7e90527f5feb805aa252"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:47:22.981434Z","signature_b64":"K9Y1xOwwjd/kTTg3zLH8vCtLGFatynByEI0V8WtCr0SaWbvu1Pz85xzbtom2byqMlqbQRddYjmTN6PZOmWVCDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5f68b70367adc32ef318d55690e355546d8f26a93a9c44edd6bffe7772b6628","last_reissued_at":"2026-07-05T06:47:22.980924Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:47:22.980924Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Split Matters: Flat Minima Methods for Improving the Performance of GNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ansgar Scherp, Nicolas Lell","submitted_at":"2023-06-15T13:29:09Z","abstract_excerpt":"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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09121","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2306.09121/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.09121","created_at":"2026-07-05T06:47:22.980982+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.09121v1","created_at":"2026-07-05T06:47:22.980982+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09121","created_at":"2026-07-05T06:47:22.980982+00:00"},{"alias_kind":"pith_short_12","alias_value":"YX3IW4BWPLOD","created_at":"2026-07-05T06:47:22.980982+00:00"},{"alias_kind":"pith_short_16","alias_value":"YX3IW4BWPLODF3ZR","created_at":"2026-07-05T06:47:22.980982+00:00"},{"alias_kind":"pith_short_8","alias_value":"YX3IW4BW","created_at":"2026-07-05T06:47:22.980982+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06151","citing_title":"Leveraging Extragradient for Effective Sharpness-Aware Minimization in Deep Learning","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV","json":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV.json","graph_json":"https://pith.science/api/pith-number/YX3IW4BWPLODF3ZRRVKWSDRVKV/graph.json","events_json":"https://pith.science/api/pith-number/YX3IW4BWPLODF3ZRRVKWSDRVKV/events.json","paper":"https://pith.science/paper/YX3IW4BW"},"agent_actions":{"view_html":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV","download_json":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV.json","view_paper":"https://pith.science/paper/YX3IW4BW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.09121&json=true","fetch_graph":"https://pith.science/api/pith-number/YX3IW4BWPLODF3ZRRVKWSDRVKV/graph.json","fetch_events":"https://pith.science/api/pith-number/YX3IW4BWPLODF3ZRRVKWSDRVKV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV/action/storage_attestation","attest_author":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV/action/author_attestation","sign_citation":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV/action/citation_signature","submit_replication":"https://pith.science/pith/YX3IW4BWPLODF3ZRRVKWSDRVKV/action/replication_record"}},"created_at":"2026-07-05T06:47:22.980982+00:00","updated_at":"2026-07-05T06:47:22.980982+00:00"}