{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:NTTX3AMYSNTEUL55YLHKHYROPG","short_pith_number":"pith:NTTX3AMY","schema_version":"1.0","canonical_sha256":"6ce77d819893664a2fbdc2cea3e22e7985868d68f39b86ea818b34af819cc4a3","source":{"kind":"arxiv","id":"2101.07706","version":1},"attestation_state":"computed","paper":{"title":"Communication-Efficient Sampling for Distributed Training of Graph Convolutional Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Masuma Akter Rumi, Peng Jiang","submitted_at":"2021-01-19T16:12:44Z","abstract_excerpt":"Training Graph Convolutional Networks (GCNs) is expensive as it needs to aggregate data recursively from neighboring nodes. To reduce the computation overhead, previous works have proposed various neighbor sampling methods that estimate the aggregation result based on a small number of sampled neighbors. Although these methods have successfully accelerated the training, they mainly focus on the single-machine setting. As real-world graphs are large, training GCNs in distributed systems is desirable. However, we found that the existing neighbor sampling methods do not work well in a distributed"},"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":"2101.07706","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-01-19T16:12:44Z","cross_cats_sorted":[],"title_canon_sha256":"03dbacaa89f4a5c62ae089a8c4ded851060bd57d718983908af857b78ff4c00f","abstract_canon_sha256":"f154a950960a5c324b57951251204c566192e3000cb42ced071856f6f364b077"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:08:00.534801Z","signature_b64":"oMacDKW2pheN7W1zoKcGdIE4Wmv6g/g1k1sVpdDQk7EugAmCeEr7A8ELySj7J6kg7v3TsrqtrSw2hQJ6GJZXBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ce77d819893664a2fbdc2cea3e22e7985868d68f39b86ea818b34af819cc4a3","last_reissued_at":"2026-07-05T02:08:00.534381Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:08:00.534381Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Communication-Efficient Sampling for Distributed Training of Graph Convolutional Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Masuma Akter Rumi, Peng Jiang","submitted_at":"2021-01-19T16:12:44Z","abstract_excerpt":"Training Graph Convolutional Networks (GCNs) is expensive as it needs to aggregate data recursively from neighboring nodes. To reduce the computation overhead, previous works have proposed various neighbor sampling methods that estimate the aggregation result based on a small number of sampled neighbors. Although these methods have successfully accelerated the training, they mainly focus on the single-machine setting. As real-world graphs are large, training GCNs in distributed systems is desirable. However, we found that the existing neighbor sampling methods do not work well in a distributed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.07706","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/2101.07706/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":"2101.07706","created_at":"2026-07-05T02:08:00.534441+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.07706v1","created_at":"2026-07-05T02:08:00.534441+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.07706","created_at":"2026-07-05T02:08:00.534441+00:00"},{"alias_kind":"pith_short_12","alias_value":"NTTX3AMYSNTE","created_at":"2026-07-05T02:08:00.534441+00:00"},{"alias_kind":"pith_short_16","alias_value":"NTTX3AMYSNTEUL55","created_at":"2026-07-05T02:08:00.534441+00:00"},{"alias_kind":"pith_short_8","alias_value":"NTTX3AMY","created_at":"2026-07-05T02:08:00.534441+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.05207","citing_title":"RapidGNN: Energy and Communication-Efficient Distributed Training on Large-Scale Graph Neural Networks","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG","json":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG.json","graph_json":"https://pith.science/api/pith-number/NTTX3AMYSNTEUL55YLHKHYROPG/graph.json","events_json":"https://pith.science/api/pith-number/NTTX3AMYSNTEUL55YLHKHYROPG/events.json","paper":"https://pith.science/paper/NTTX3AMY"},"agent_actions":{"view_html":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG","download_json":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG.json","view_paper":"https://pith.science/paper/NTTX3AMY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.07706&json=true","fetch_graph":"https://pith.science/api/pith-number/NTTX3AMYSNTEUL55YLHKHYROPG/graph.json","fetch_events":"https://pith.science/api/pith-number/NTTX3AMYSNTEUL55YLHKHYROPG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG/action/storage_attestation","attest_author":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG/action/author_attestation","sign_citation":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG/action/citation_signature","submit_replication":"https://pith.science/pith/NTTX3AMYSNTEUL55YLHKHYROPG/action/replication_record"}},"created_at":"2026-07-05T02:08:00.534441+00:00","updated_at":"2026-07-05T02:08:00.534441+00:00"}