{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:CCXYJUEDM5NMGWFD5UK3SSDCWS","short_pith_number":"pith:CCXYJUED","schema_version":"1.0","canonical_sha256":"10af84d083675ac358a3ed15b94862b4b39e01875ffce7f814a5f5aaa9c5394f","source":{"kind":"arxiv","id":"1905.04374","version":2},"attestation_state":"computed","paper":{"title":"Fast and Robust Distributed Learning in High Dimension","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CR","cs.LG","stat.ML"],"primary_cat":"cs.DC","authors_text":"El-Mahdi El-Mhamdi, Rachid Guerraoui, S\\'ebastien Rouault","submitted_at":"2019-05-05T16:41:25Z","abstract_excerpt":"Could a gradient aggregation rule (GAR) for distributed machine learning be both robust and fast? This paper answers by the affirmative through multi-Bulyan. Given $n$ workers, $f$ of which are arbitrary malicious (Byzantine) and $m=n-f$ are not, we prove that multi-Bulyan can ensure a strong form of Byzantine resilience, as well as an ${\\frac{m}{n}}$ slowdown, compared to averaging, the fastest (but non Byzantine resilient) rule for distributed machine learning. When $m \\approx n$ (almost all workers are correct), multi-Bulyan reaches the speed of averaging. We also prove that multi-Bulyan's "},"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":"1905.04374","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2019-05-05T16:41:25Z","cross_cats_sorted":["cs.CR","cs.LG","stat.ML"],"title_canon_sha256":"7641086064dc2023c2bd86062b1ba539951da6669d75a5fe9bb5998d95e5cad4","abstract_canon_sha256":"fb682c6da1f3313da7a3b9793c9f7eeec2d652e3c68cdeb2d8fa102efbb5d35d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:13:01.520893Z","signature_b64":"waoH0AMJe3NIvJtzHqrgc73A6TK2O1CNK57ffNsdR5ahY7SusI0kv7D6cOH0weTDrqMPqzKUgPpqtFK47t0ZDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10af84d083675ac358a3ed15b94862b4b39e01875ffce7f814a5f5aaa9c5394f","last_reissued_at":"2026-07-05T02:13:01.519481Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:13:01.519481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast and Robust Distributed Learning in High Dimension","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CR","cs.LG","stat.ML"],"primary_cat":"cs.DC","authors_text":"El-Mahdi El-Mhamdi, Rachid Guerraoui, S\\'ebastien Rouault","submitted_at":"2019-05-05T16:41:25Z","abstract_excerpt":"Could a gradient aggregation rule (GAR) for distributed machine learning be both robust and fast? This paper answers by the affirmative through multi-Bulyan. Given $n$ workers, $f$ of which are arbitrary malicious (Byzantine) and $m=n-f$ are not, we prove that multi-Bulyan can ensure a strong form of Byzantine resilience, as well as an ${\\frac{m}{n}}$ slowdown, compared to averaging, the fastest (but non Byzantine resilient) rule for distributed machine learning. When $m \\approx n$ (almost all workers are correct), multi-Bulyan reaches the speed of averaging. We also prove that multi-Bulyan's "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.04374","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/1905.04374/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":"1905.04374","created_at":"2026-07-05T02:13:01.519673+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.04374v2","created_at":"2026-07-05T02:13:01.519673+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.04374","created_at":"2026-07-05T02:13:01.519673+00:00"},{"alias_kind":"pith_short_12","alias_value":"CCXYJUEDM5NM","created_at":"2026-07-05T02:13:01.519673+00:00"},{"alias_kind":"pith_short_16","alias_value":"CCXYJUEDM5NMGWFD","created_at":"2026-07-05T02:13:01.519673+00:00"},{"alias_kind":"pith_short_8","alias_value":"CCXYJUED","created_at":"2026-07-05T02:13:01.519673+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS","json":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS.json","graph_json":"https://pith.science/api/pith-number/CCXYJUEDM5NMGWFD5UK3SSDCWS/graph.json","events_json":"https://pith.science/api/pith-number/CCXYJUEDM5NMGWFD5UK3SSDCWS/events.json","paper":"https://pith.science/paper/CCXYJUED"},"agent_actions":{"view_html":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS","download_json":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS.json","view_paper":"https://pith.science/paper/CCXYJUED","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.04374&json=true","fetch_graph":"https://pith.science/api/pith-number/CCXYJUEDM5NMGWFD5UK3SSDCWS/graph.json","fetch_events":"https://pith.science/api/pith-number/CCXYJUEDM5NMGWFD5UK3SSDCWS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS/action/storage_attestation","attest_author":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS/action/author_attestation","sign_citation":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS/action/citation_signature","submit_replication":"https://pith.science/pith/CCXYJUEDM5NMGWFD5UK3SSDCWS/action/replication_record"}},"created_at":"2026-07-05T02:13:01.519673+00:00","updated_at":"2026-07-05T02:13:01.519673+00:00"}