{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F7DRTRRTTYQF2Q7H2WSXAP3VGM","short_pith_number":"pith:F7DRTRRT","schema_version":"1.0","canonical_sha256":"2fc719c6339e205d43e7d5a5703f7533176b3a7f31742fc9919acbe36b18b52c","source":{"kind":"arxiv","id":"2411.07061","version":1},"attestation_state":"computed","paper":{"title":"General framework for online-to-nonconvex conversion: Schedule-free SGD is also effective for nonconvex optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashok Cutkosky, Gagik Magakyan, Kwangjun Ahn","submitted_at":"2024-11-11T15:25:48Z","abstract_excerpt":"This work investigates the effectiveness of schedule-free methods, developed by A. Defazio et al. (NeurIPS 2024), in nonconvex optimization settings, inspired by their remarkable empirical success in training neural networks. Specifically, we show that schedule-free SGD achieves optimal iteration complexity for nonsmooth, nonconvex optimization problems. Our proof begins with the development of a general framework for online-to-nonconvex conversion, which converts a given online learning algorithm into an optimization algorithm for nonconvex losses. Our general framework not only recovers exis"},"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":"2411.07061","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-11T15:25:48Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"12ebc92fa80170f0a07a3cc006e94e7f7a6ea97230747bfd94c2bfe12fb4d094","abstract_canon_sha256":"939be9720f81083dac1589c59d7c68e011506390d14cbe1c36ad8b5c792e6bf6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:56.053088Z","signature_b64":"lI8fUYpYka5OVWWMMHn//Vsq4o62YmdowpPyZf5IgWG/MICwKIoQSLrRj+Etr68axLb8bsy0T9ZAWU561+kRAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fc719c6339e205d43e7d5a5703f7533176b3a7f31742fc9919acbe36b18b52c","last_reissued_at":"2026-07-05T09:33:56.052663Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:56.052663Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"General framework for online-to-nonconvex conversion: Schedule-free SGD is also effective for nonconvex optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashok Cutkosky, Gagik Magakyan, Kwangjun Ahn","submitted_at":"2024-11-11T15:25:48Z","abstract_excerpt":"This work investigates the effectiveness of schedule-free methods, developed by A. Defazio et al. (NeurIPS 2024), in nonconvex optimization settings, inspired by their remarkable empirical success in training neural networks. Specifically, we show that schedule-free SGD achieves optimal iteration complexity for nonsmooth, nonconvex optimization problems. Our proof begins with the development of a general framework for online-to-nonconvex conversion, which converts a given online learning algorithm into an optimization algorithm for nonconvex losses. Our general framework not only recovers exis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.07061","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/2411.07061/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":"2411.07061","created_at":"2026-07-05T09:33:56.052714+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.07061v1","created_at":"2026-07-05T09:33:56.052714+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.07061","created_at":"2026-07-05T09:33:56.052714+00:00"},{"alias_kind":"pith_short_12","alias_value":"F7DRTRRTTYQF","created_at":"2026-07-05T09:33:56.052714+00:00"},{"alias_kind":"pith_short_16","alias_value":"F7DRTRRTTYQF2Q7H","created_at":"2026-07-05T09:33:56.052714+00:00"},{"alias_kind":"pith_short_8","alias_value":"F7DRTRRT","created_at":"2026-07-05T09:33:56.052714+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21433","citing_title":"Central limit theorem for the averaged Adam optimizer","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM","json":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM.json","graph_json":"https://pith.science/api/pith-number/F7DRTRRTTYQF2Q7H2WSXAP3VGM/graph.json","events_json":"https://pith.science/api/pith-number/F7DRTRRTTYQF2Q7H2WSXAP3VGM/events.json","paper":"https://pith.science/paper/F7DRTRRT"},"agent_actions":{"view_html":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM","download_json":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM.json","view_paper":"https://pith.science/paper/F7DRTRRT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.07061&json=true","fetch_graph":"https://pith.science/api/pith-number/F7DRTRRTTYQF2Q7H2WSXAP3VGM/graph.json","fetch_events":"https://pith.science/api/pith-number/F7DRTRRTTYQF2Q7H2WSXAP3VGM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM/action/storage_attestation","attest_author":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM/action/author_attestation","sign_citation":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM/action/citation_signature","submit_replication":"https://pith.science/pith/F7DRTRRTTYQF2Q7H2WSXAP3VGM/action/replication_record"}},"created_at":"2026-07-05T09:33:56.052714+00:00","updated_at":"2026-07-05T09:33:56.052714+00:00"}