{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F6NTNIAKEMEZWBGRUUOQDRWLRT","short_pith_number":"pith:F6NTNIAK","schema_version":"1.0","canonical_sha256":"2f9b36a00a23099b04d1a51d01c6cb8cc5a5c8be6c59ea4143fc094f59e1459a","source":{"kind":"arxiv","id":"2410.12918","version":2},"attestation_state":"computed","paper":{"title":"Boosting Asynchronous Decentralized Learning with Model Fragmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Alexis Marouani, Anne-Marie Kermarrec, Martijn de Vos, Rafael Pires, Rishi Sharma, Sayan Biswas","submitted_at":"2024-10-16T18:03:52Z","abstract_excerpt":"Decentralized learning (DL) is an emerging technique that allows nodes on the web to collaboratively train machine learning models without sharing raw data. Dealing with stragglers, i.e., nodes with slower compute or communication than others, is a key challenge in DL. We present DivShare, a novel asynchronous DL algorithm that achieves fast model convergence in the presence of communication stragglers. DivShare achieves this by having nodes fragment their models into parameter subsets and send, in parallel to computation, each subset to a random sample of other nodes instead of sequentially e"},"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":"2410.12918","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-10-16T18:03:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"00d8b2dd33fb5d9e30559af3313e6b0d7741e6363a12b7361229bca31fad096e","abstract_canon_sha256":"dded0d762494be9379cb33e7818da84b9c52763e3f1340d03ab1046ca47560c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:53.907669Z","signature_b64":"t/U/laAOf3OBMzjHpWYqXAcwz0yyd364kA/8rL7qTqMNeciYb+Eaoxzw6IBB1ZdDzTbhQxVoPPp+n50y6eF2Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f9b36a00a23099b04d1a51d01c6cb8cc5a5c8be6c59ea4143fc094f59e1459a","last_reissued_at":"2026-07-05T10:08:53.907217Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:53.907217Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Boosting Asynchronous Decentralized Learning with Model Fragmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Alexis Marouani, Anne-Marie Kermarrec, Martijn de Vos, Rafael Pires, Rishi Sharma, Sayan Biswas","submitted_at":"2024-10-16T18:03:52Z","abstract_excerpt":"Decentralized learning (DL) is an emerging technique that allows nodes on the web to collaboratively train machine learning models without sharing raw data. Dealing with stragglers, i.e., nodes with slower compute or communication than others, is a key challenge in DL. We present DivShare, a novel asynchronous DL algorithm that achieves fast model convergence in the presence of communication stragglers. DivShare achieves this by having nodes fragment their models into parameter subsets and send, in parallel to computation, each subset to a random sample of other nodes instead of sequentially e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.12918","kind":"arxiv","version":2},"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/2410.12918/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":"2410.12918","created_at":"2026-07-05T10:08:53.907272+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.12918v2","created_at":"2026-07-05T10:08:53.907272+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.12918","created_at":"2026-07-05T10:08:53.907272+00:00"},{"alias_kind":"pith_short_12","alias_value":"F6NTNIAKEMEZ","created_at":"2026-07-05T10:08:53.907272+00:00"},{"alias_kind":"pith_short_16","alias_value":"F6NTNIAKEMEZWBGR","created_at":"2026-07-05T10:08:53.907272+00:00"},{"alias_kind":"pith_short_8","alias_value":"F6NTNIAK","created_at":"2026-07-05T10:08:53.907272+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.10911","citing_title":"NoLoCo: No-all-reduce Low Communication Training Method for Large Models","ref_index":2020,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT","json":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT.json","graph_json":"https://pith.science/api/pith-number/F6NTNIAKEMEZWBGRUUOQDRWLRT/graph.json","events_json":"https://pith.science/api/pith-number/F6NTNIAKEMEZWBGRUUOQDRWLRT/events.json","paper":"https://pith.science/paper/F6NTNIAK"},"agent_actions":{"view_html":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT","download_json":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT.json","view_paper":"https://pith.science/paper/F6NTNIAK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.12918&json=true","fetch_graph":"https://pith.science/api/pith-number/F6NTNIAKEMEZWBGRUUOQDRWLRT/graph.json","fetch_events":"https://pith.science/api/pith-number/F6NTNIAKEMEZWBGRUUOQDRWLRT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT/action/storage_attestation","attest_author":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT/action/author_attestation","sign_citation":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT/action/citation_signature","submit_replication":"https://pith.science/pith/F6NTNIAKEMEZWBGRUUOQDRWLRT/action/replication_record"}},"created_at":"2026-07-05T10:08:53.907272+00:00","updated_at":"2026-07-05T10:08:53.907272+00:00"}