{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YIGYMOWHEKT234TQELXBM5PE7E","short_pith_number":"pith:YIGYMOWH","schema_version":"1.0","canonical_sha256":"c20d863ac722a7adf27022ee1675e4f9009be1f7d9a6f605f6ba06619db93d02","source":{"kind":"arxiv","id":"2310.10953","version":1},"attestation_state":"computed","paper":{"title":"A Local Graph Limits Perspective on Sampling-Based GNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Amin Saberi, Luana Ruiz, Yeganeh Alimohammadi","submitted_at":"2023-10-17T02:58:49Z","abstract_excerpt":"We propose a theoretical framework for training Graph Neural Networks (GNNs) on large input graphs via training on small, fixed-size sampled subgraphs. This framework is applicable to a wide range of models, including popular sampling-based GNNs, such as GraphSAGE and FastGCN. Leveraging the theory of graph local limits, we prove that, under mild assumptions, parameters learned from training sampling-based GNNs on small samples of a large input graph are within an $\\epsilon$-neighborhood of the outcome of training the same architecture on the whole graph. We derive bounds on the number of samp"},"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":"2310.10953","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-17T02:58:49Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"236b506f16b64ae738cef40bf91a5a782a021d89bef1b2e88477f3da8353a5ec","abstract_canon_sha256":"66e24c8ab4ef9f686eb22f1a5cb9262cb83bb587da829cb334150462a010443c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:01:29.956237Z","signature_b64":"QGXdUO4k9saaevgQNc5VlPjGfpd02eQeV7u/xK5AgeWrnMVppdSBKSQCaY6kFX965wgQ5+G7MxHO7kC1isxtBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c20d863ac722a7adf27022ee1675e4f9009be1f7d9a6f605f6ba06619db93d02","last_reissued_at":"2026-07-05T07:01:29.955774Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:01:29.955774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Local Graph Limits Perspective on Sampling-Based GNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Amin Saberi, Luana Ruiz, Yeganeh Alimohammadi","submitted_at":"2023-10-17T02:58:49Z","abstract_excerpt":"We propose a theoretical framework for training Graph Neural Networks (GNNs) on large input graphs via training on small, fixed-size sampled subgraphs. This framework is applicable to a wide range of models, including popular sampling-based GNNs, such as GraphSAGE and FastGCN. Leveraging the theory of graph local limits, we prove that, under mild assumptions, parameters learned from training sampling-based GNNs on small samples of a large input graph are within an $\\epsilon$-neighborhood of the outcome of training the same architecture on the whole graph. We derive bounds on the number of samp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.10953","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/2310.10953/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":"2310.10953","created_at":"2026-07-05T07:01:29.955833+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.10953v1","created_at":"2026-07-05T07:01:29.955833+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.10953","created_at":"2026-07-05T07:01:29.955833+00:00"},{"alias_kind":"pith_short_12","alias_value":"YIGYMOWHEKT2","created_at":"2026-07-05T07:01:29.955833+00:00"},{"alias_kind":"pith_short_16","alias_value":"YIGYMOWHEKT234TQ","created_at":"2026-07-05T07:01:29.955833+00:00"},{"alias_kind":"pith_short_8","alias_value":"YIGYMOWH","created_at":"2026-07-05T07:01:29.955833+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01161","citing_title":"AcOrch: Accelerating Sampling-based GNN Training under CPU-NPU Heterogeneous Environments","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E","json":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E.json","graph_json":"https://pith.science/api/pith-number/YIGYMOWHEKT234TQELXBM5PE7E/graph.json","events_json":"https://pith.science/api/pith-number/YIGYMOWHEKT234TQELXBM5PE7E/events.json","paper":"https://pith.science/paper/YIGYMOWH"},"agent_actions":{"view_html":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E","download_json":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E.json","view_paper":"https://pith.science/paper/YIGYMOWH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.10953&json=true","fetch_graph":"https://pith.science/api/pith-number/YIGYMOWHEKT234TQELXBM5PE7E/graph.json","fetch_events":"https://pith.science/api/pith-number/YIGYMOWHEKT234TQELXBM5PE7E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E/action/storage_attestation","attest_author":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E/action/author_attestation","sign_citation":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E/action/citation_signature","submit_replication":"https://pith.science/pith/YIGYMOWHEKT234TQELXBM5PE7E/action/replication_record"}},"created_at":"2026-07-05T07:01:29.955833+00:00","updated_at":"2026-07-05T07:01:29.955833+00:00"}