{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:R6Z6YUXPG27WT2LT2DHSZROPJZ","short_pith_number":"pith:R6Z6YUXP","schema_version":"1.0","canonical_sha256":"8fb3ec52ef36bf69e973d0cf2cc5cf4e755f7da0151415609a98aa04d99b178a","source":{"kind":"arxiv","id":"2505.10889","version":1},"attestation_state":"computed","paper":{"title":"Convergence Analysis of the Last Iterate in Distributed Stochastic Gradient Descent with Momentum","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"math.OC","authors_text":"Bo Zhang, Difei Cheng, Hong Qiao, Ruinan Jin","submitted_at":"2025-05-16T05:52:12Z","abstract_excerpt":"Distributed stochastic gradient methods are widely used to preserve data privacy and ensure scalability in large-scale learning tasks. While existing theory on distributed momentum Stochastic Gradient Descent (mSGD) mainly focuses on time-averaged convergence, the more practical last-iterate convergence remains underexplored. In this work, we analyze the last-iterate convergence behavior of distributed mSGD in non-convex settings under the classical Robbins-Monro step-size schedule. We prove both almost sure convergence and $L_2$ convergence of the last iterate, and derive convergence rates. W"},"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":"2505.10889","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2025-05-16T05:52:12Z","cross_cats_sorted":[],"title_canon_sha256":"70026fa5b190633fdb858134f6986b51eee2a84490c8de1178d800a47b6ee07f","abstract_canon_sha256":"244cca16630bba5628038f75626558229885c15716032f1d93c18d7e06430062"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:07.700214Z","signature_b64":"DeIesMAUlwCzETP8cdw5DtFOzAOOt7PqsFI2DAGkVsnXwAtIEpCZ/zcki/Jjx1eDzTxdmK9Wj5LdncndWeyxCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8fb3ec52ef36bf69e973d0cf2cc5cf4e755f7da0151415609a98aa04d99b178a","last_reissued_at":"2026-07-05T11:04:07.699796Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:07.699796Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Convergence Analysis of the Last Iterate in Distributed Stochastic Gradient Descent with Momentum","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"math.OC","authors_text":"Bo Zhang, Difei Cheng, Hong Qiao, Ruinan Jin","submitted_at":"2025-05-16T05:52:12Z","abstract_excerpt":"Distributed stochastic gradient methods are widely used to preserve data privacy and ensure scalability in large-scale learning tasks. While existing theory on distributed momentum Stochastic Gradient Descent (mSGD) mainly focuses on time-averaged convergence, the more practical last-iterate convergence remains underexplored. In this work, we analyze the last-iterate convergence behavior of distributed mSGD in non-convex settings under the classical Robbins-Monro step-size schedule. We prove both almost sure convergence and $L_2$ convergence of the last iterate, and derive convergence rates. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.10889","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/2505.10889/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":"2505.10889","created_at":"2026-07-05T11:04:07.699858+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.10889v1","created_at":"2026-07-05T11:04:07.699858+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.10889","created_at":"2026-07-05T11:04:07.699858+00:00"},{"alias_kind":"pith_short_12","alias_value":"R6Z6YUXPG27W","created_at":"2026-07-05T11:04:07.699858+00:00"},{"alias_kind":"pith_short_16","alias_value":"R6Z6YUXPG27WT2LT","created_at":"2026-07-05T11:04:07.699858+00:00"},{"alias_kind":"pith_short_8","alias_value":"R6Z6YUXP","created_at":"2026-07-05T11:04:07.699858+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/R6Z6YUXPG27WT2LT2DHSZROPJZ","json":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ.json","graph_json":"https://pith.science/api/pith-number/R6Z6YUXPG27WT2LT2DHSZROPJZ/graph.json","events_json":"https://pith.science/api/pith-number/R6Z6YUXPG27WT2LT2DHSZROPJZ/events.json","paper":"https://pith.science/paper/R6Z6YUXP"},"agent_actions":{"view_html":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ","download_json":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ.json","view_paper":"https://pith.science/paper/R6Z6YUXP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.10889&json=true","fetch_graph":"https://pith.science/api/pith-number/R6Z6YUXPG27WT2LT2DHSZROPJZ/graph.json","fetch_events":"https://pith.science/api/pith-number/R6Z6YUXPG27WT2LT2DHSZROPJZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ/action/storage_attestation","attest_author":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ/action/author_attestation","sign_citation":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ/action/citation_signature","submit_replication":"https://pith.science/pith/R6Z6YUXPG27WT2LT2DHSZROPJZ/action/replication_record"}},"created_at":"2026-07-05T11:04:07.699858+00:00","updated_at":"2026-07-05T11:04:07.699858+00:00"}