{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z3N5ETRBFYPENDKSQSVJ5UDYUP","short_pith_number":"pith:Z3N5ETRB","schema_version":"1.0","canonical_sha256":"cedbd24e212e1e468d5284aa9ed078a3e2cd89ea2d6beae759874f04433b79df","source":{"kind":"arxiv","id":"2505.23081","version":2},"attestation_state":"computed","paper":{"title":"Gradient Methods with Online Scaling Part I. Theoretical Foundations","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Madeleine Udell, Wenzhi Gao, Ya-Chi Chu, Yinyu Ye","submitted_at":"2025-05-29T04:35:21Z","abstract_excerpt":"This paper establishes the theoretical foundations of the online scaled gradient methods (OSGM), a framework that utilizes online learning to adapt stepsizes and provably accelerate first-order methods. OSGM quantifies the effectiveness of a stepsize by a feedback function motivated from a convergence measure and uses the feedback to adjust the stepsize through an online learning algorithm. Consequently, instantiations of OSGM achieve convergence rates that are asymptotically no worse than the optimal stepsize. OSGM yields desirable convergence guarantees on smooth convex problems, including 1"},"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":"2505.23081","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"math.OC","submitted_at":"2025-05-29T04:35:21Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"2259afa251c2413f79a04185eb06a057a166caea396229befbfaf5a6adaa7692","abstract_canon_sha256":"80cc776fc9bd88b4ff6ff8c19306b2dd8b2b6c63cd86e01af6fa715850ea14d7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:17.165616Z","signature_b64":"CXFPUdM67qBwGAB/GrUBUrzub+j7sRF4KfmItNgw6c2PkOTTDnWmmlbL64vCVXNX79D9uBiAXlWFR8KDUL+1BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cedbd24e212e1e468d5284aa9ed078a3e2cd89ea2d6beae759874f04433b79df","last_reissued_at":"2026-07-05T12:05:17.165079Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:17.165079Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gradient Methods with Online Scaling Part I. Theoretical Foundations","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Madeleine Udell, Wenzhi Gao, Ya-Chi Chu, Yinyu Ye","submitted_at":"2025-05-29T04:35:21Z","abstract_excerpt":"This paper establishes the theoretical foundations of the online scaled gradient methods (OSGM), a framework that utilizes online learning to adapt stepsizes and provably accelerate first-order methods. OSGM quantifies the effectiveness of a stepsize by a feedback function motivated from a convergence measure and uses the feedback to adjust the stepsize through an online learning algorithm. Consequently, instantiations of OSGM achieve convergence rates that are asymptotically no worse than the optimal stepsize. OSGM yields desirable convergence guarantees on smooth convex problems, including 1"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23081","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/2505.23081/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":"2505.23081","created_at":"2026-07-05T12:05:17.165145+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.23081v2","created_at":"2026-07-05T12:05:17.165145+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23081","created_at":"2026-07-05T12:05:17.165145+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z3N5ETRBFYPE","created_at":"2026-07-05T12:05:17.165145+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z3N5ETRBFYPENDKS","created_at":"2026-07-05T12:05:17.165145+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z3N5ETRB","created_at":"2026-07-05T12:05:17.165145+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.05288","citing_title":"Learning to accelerate distributed ADMM using graph neural networks","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP","json":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP.json","graph_json":"https://pith.science/api/pith-number/Z3N5ETRBFYPENDKSQSVJ5UDYUP/graph.json","events_json":"https://pith.science/api/pith-number/Z3N5ETRBFYPENDKSQSVJ5UDYUP/events.json","paper":"https://pith.science/paper/Z3N5ETRB"},"agent_actions":{"view_html":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP","download_json":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP.json","view_paper":"https://pith.science/paper/Z3N5ETRB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.23081&json=true","fetch_graph":"https://pith.science/api/pith-number/Z3N5ETRBFYPENDKSQSVJ5UDYUP/graph.json","fetch_events":"https://pith.science/api/pith-number/Z3N5ETRBFYPENDKSQSVJ5UDYUP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP/action/storage_attestation","attest_author":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP/action/author_attestation","sign_citation":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP/action/citation_signature","submit_replication":"https://pith.science/pith/Z3N5ETRBFYPENDKSQSVJ5UDYUP/action/replication_record"}},"created_at":"2026-07-05T12:05:17.165145+00:00","updated_at":"2026-07-05T12:05:17.165145+00:00"}