{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:H24KT3KL7QGRDP6YLRENV4ICZR","short_pith_number":"pith:H24KT3KL","schema_version":"1.0","canonical_sha256":"3eb8a9ed4bfc0d11bfd85c48daf102cc4740617e500fa83d0a3dacfdeda04a78","source":{"kind":"arxiv","id":"2112.03968","version":1},"attestation_state":"computed","paper":{"title":"Learning Theory Can (Sometimes) Explain Generalisation in Graph Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Debarghya Ghoshdastidar, Leena Chennuru Vankadara, Pascal Mattia Esser","submitted_at":"2021-12-07T20:06:23Z","abstract_excerpt":"In recent years, several results in the supervised learning setting suggested that classical statistical learning-theoretic measures, such as VC dimension, do not adequately explain the performance of deep learning models which prompted a slew of work in the infinite-width and iteration regimes. However, there is little theoretical explanation for the success of neural networks beyond the supervised setting. In this paper we argue that, under some distributional assumptions, classical learning-theoretic measures can sufficiently explain generalization for graph neural networks in the transduct"},"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":"2112.03968","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-07T20:06:23Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"d886dd4f486ba8631410465d679fb50528953ba8e336138a00e2f9cf84cf5ecd","abstract_canon_sha256":"bab1bc2f342799bf3510eb15773afcdeaafd505ee3616b9805487704a6cf2b84"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:38:52.418986Z","signature_b64":"SeNR0v9b7deSWW+hvPSTAJk8FoaaRS6A8HtDDwgfNIJPFierezQj/U0EYvx5/DHkOGv71lxOtlxn772gvVT2Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3eb8a9ed4bfc0d11bfd85c48daf102cc4740617e500fa83d0a3dacfdeda04a78","last_reissued_at":"2026-07-05T03:38:52.418598Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:38:52.418598Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Theory Can (Sometimes) Explain Generalisation in Graph Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Debarghya Ghoshdastidar, Leena Chennuru Vankadara, Pascal Mattia Esser","submitted_at":"2021-12-07T20:06:23Z","abstract_excerpt":"In recent years, several results in the supervised learning setting suggested that classical statistical learning-theoretic measures, such as VC dimension, do not adequately explain the performance of deep learning models which prompted a slew of work in the infinite-width and iteration regimes. However, there is little theoretical explanation for the success of neural networks beyond the supervised setting. In this paper we argue that, under some distributional assumptions, classical learning-theoretic measures can sufficiently explain generalization for graph neural networks in the transduct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.03968","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/2112.03968/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":"2112.03968","created_at":"2026-07-05T03:38:52.418651+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.03968v1","created_at":"2026-07-05T03:38:52.418651+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.03968","created_at":"2026-07-05T03:38:52.418651+00:00"},{"alias_kind":"pith_short_12","alias_value":"H24KT3KL7QGR","created_at":"2026-07-05T03:38:52.418651+00:00"},{"alias_kind":"pith_short_16","alias_value":"H24KT3KL7QGRDP6Y","created_at":"2026-07-05T03:38:52.418651+00:00"},{"alias_kind":"pith_short_8","alias_value":"H24KT3KL","created_at":"2026-07-05T03:38:52.418651+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.11162","citing_title":"RPN 2: On Interdependence Function Learning Towards Unifying and Advancing CNN, RNN, GNN, and Transformer","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR","json":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR.json","graph_json":"https://pith.science/api/pith-number/H24KT3KL7QGRDP6YLRENV4ICZR/graph.json","events_json":"https://pith.science/api/pith-number/H24KT3KL7QGRDP6YLRENV4ICZR/events.json","paper":"https://pith.science/paper/H24KT3KL"},"agent_actions":{"view_html":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR","download_json":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR.json","view_paper":"https://pith.science/paper/H24KT3KL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.03968&json=true","fetch_graph":"https://pith.science/api/pith-number/H24KT3KL7QGRDP6YLRENV4ICZR/graph.json","fetch_events":"https://pith.science/api/pith-number/H24KT3KL7QGRDP6YLRENV4ICZR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR/action/storage_attestation","attest_author":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR/action/author_attestation","sign_citation":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR/action/citation_signature","submit_replication":"https://pith.science/pith/H24KT3KL7QGRDP6YLRENV4ICZR/action/replication_record"}},"created_at":"2026-07-05T03:38:52.418651+00:00","updated_at":"2026-07-05T03:38:52.418651+00:00"}