{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IMDQZIJIPEIPN3VMS6VZNB2PLA","short_pith_number":"pith:IMDQZIJI","schema_version":"1.0","canonical_sha256":"43070ca1287910f6eeac97ab96874f58378c875a2851314b9d8121f689966b1d","source":{"kind":"arxiv","id":"2508.12327","version":1},"attestation_state":"computed","paper":{"title":"Convergence Analysis of the Lion Optimizer in Centralized and Distributed Settings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Lijun Zhang, Wei Jiang","submitted_at":"2025-08-17T10:54:01Z","abstract_excerpt":"In this paper, we analyze the convergence properties of the Lion optimizer. First, we establish that the Lion optimizer attains a convergence rate of $\\mathcal{O}(d^{1/2}T^{-1/4})$ under standard assumptions, where $d$ denotes the problem dimension and $T$ is the iteration number. To further improve this rate, we introduce the Lion optimizer with variance reduction, resulting in an enhanced convergence rate of $\\mathcal{O}(d^{1/2}T^{-1/3})$. We then analyze in distributed settings, where the standard and variance reduced version of the distributed Lion can obtain the convergence rates of $\\mat"},"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":"2508.12327","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-17T10:54:01Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"d2d4c58cedffe6b371d82f64d5f9a07e8a025dca6d8895554964cce16ab4dfbc","abstract_canon_sha256":"57e7c5ed943ccd24ac66c11be57fe7fd8973d2dee1c19887323cfd8a7a1984a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:03.617898Z","signature_b64":"aiHWbo4ROnGJgsLLYVmELkmrN+DeqUKpgiCmW1JomS80PVDJlh7RRqCQlQYqC+6Go30rQ7+vZZ4Dqsh65Eq9CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43070ca1287910f6eeac97ab96874f58378c875a2851314b9d8121f689966b1d","last_reissued_at":"2026-07-05T11:55:03.617421Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:03.617421Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Convergence Analysis of the Lion Optimizer in Centralized and Distributed Settings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Lijun Zhang, Wei Jiang","submitted_at":"2025-08-17T10:54:01Z","abstract_excerpt":"In this paper, we analyze the convergence properties of the Lion optimizer. First, we establish that the Lion optimizer attains a convergence rate of $\\mathcal{O}(d^{1/2}T^{-1/4})$ under standard assumptions, where $d$ denotes the problem dimension and $T$ is the iteration number. To further improve this rate, we introduce the Lion optimizer with variance reduction, resulting in an enhanced convergence rate of $\\mathcal{O}(d^{1/2}T^{-1/3})$. We then analyze in distributed settings, where the standard and variance reduced version of the distributed Lion can obtain the convergence rates of $\\mat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.12327","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/2508.12327/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":"2508.12327","created_at":"2026-07-05T11:55:03.617480+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.12327v1","created_at":"2026-07-05T11:55:03.617480+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.12327","created_at":"2026-07-05T11:55:03.617480+00:00"},{"alias_kind":"pith_short_12","alias_value":"IMDQZIJIPEIP","created_at":"2026-07-05T11:55:03.617480+00:00"},{"alias_kind":"pith_short_16","alias_value":"IMDQZIJIPEIPN3VM","created_at":"2026-07-05T11:55:03.617480+00:00"},{"alias_kind":"pith_short_8","alias_value":"IMDQZIJI","created_at":"2026-07-05T11:55:03.617480+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06615","citing_title":"When and Why SignSGD Outperforms SGD: A Theoretical Study Based on $\\ell_1$-norm Lower Bounds","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05577","citing_title":"Accelerating LMO-Based Optimization via Implicit Gradient Transport","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15416","citing_title":"StoSignSGD: Unbiased Structural Stochasticity Fixes SignSGD for Training Large Language Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14587","citing_title":"CLion: Efficient Cautious Lion Optimizer with Enhanced Generalization","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA","json":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA.json","graph_json":"https://pith.science/api/pith-number/IMDQZIJIPEIPN3VMS6VZNB2PLA/graph.json","events_json":"https://pith.science/api/pith-number/IMDQZIJIPEIPN3VMS6VZNB2PLA/events.json","paper":"https://pith.science/paper/IMDQZIJI"},"agent_actions":{"view_html":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA","download_json":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA.json","view_paper":"https://pith.science/paper/IMDQZIJI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.12327&json=true","fetch_graph":"https://pith.science/api/pith-number/IMDQZIJIPEIPN3VMS6VZNB2PLA/graph.json","fetch_events":"https://pith.science/api/pith-number/IMDQZIJIPEIPN3VMS6VZNB2PLA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA/action/storage_attestation","attest_author":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA/action/author_attestation","sign_citation":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA/action/citation_signature","submit_replication":"https://pith.science/pith/IMDQZIJIPEIPN3VMS6VZNB2PLA/action/replication_record"}},"created_at":"2026-07-05T11:55:03.617480+00:00","updated_at":"2026-07-05T11:55:03.617480+00:00"}