{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:QGCOMYEAW6KYC5FNYP57NT23SL","short_pith_number":"pith:QGCOMYEA","schema_version":"1.0","canonical_sha256":"8184e66080b7958174adc3fbf6cf5b92c5f49620b635938f7df30acdab11e632","source":{"kind":"arxiv","id":"2006.10311","version":3},"attestation_state":"computed","paper":{"title":"SGD for Structured Nonconvex Functions: Learning Rates, Minibatching and Interpolation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Nicolas Loizou, Othmane Sebbouh, Robert M. Gower","submitted_at":"2020-06-18T07:05:56Z","abstract_excerpt":"Stochastic Gradient Descent (SGD) is being used routinely for optimizing non-convex functions. Yet, the standard convergence theory for SGD in the smooth non-convex setting gives a slow sublinear convergence to a stationary point. In this work, we provide several convergence theorems for SGD showing convergence to a global minimum for non-convex problems satisfying some extra structural assumptions. In particular, we focus on two large classes of structured non-convex functions: (i) Quasar (Strongly) Convex functions (a generalization of convex functions) and (ii) functions satisfying the Poly"},"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":"2006.10311","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2020-06-18T07:05:56Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"0227cf9f6b996f897306d89fa7b9f89ae728d3a2ee57e9efac1fbda8b510cc1b","abstract_canon_sha256":"0975b96d1860f22a97998eea1d80ef98d238d1ff66410f624e9cc1abfb51d788"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:25:10.347227Z","signature_b64":"szZkKn/Qj3n4DRzpiRfB8Wd3rmig0kZhksY/2spZ0kKPvH3r/XJha8loePwgXlm5qKbE1lEJzv01xShCOp6aDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8184e66080b7958174adc3fbf6cf5b92c5f49620b635938f7df30acdab11e632","last_reissued_at":"2026-07-05T02:25:10.346752Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:25:10.346752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SGD for Structured Nonconvex Functions: Learning Rates, Minibatching and Interpolation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Nicolas Loizou, Othmane Sebbouh, Robert M. Gower","submitted_at":"2020-06-18T07:05:56Z","abstract_excerpt":"Stochastic Gradient Descent (SGD) is being used routinely for optimizing non-convex functions. Yet, the standard convergence theory for SGD in the smooth non-convex setting gives a slow sublinear convergence to a stationary point. In this work, we provide several convergence theorems for SGD showing convergence to a global minimum for non-convex problems satisfying some extra structural assumptions. In particular, we focus on two large classes of structured non-convex functions: (i) Quasar (Strongly) Convex functions (a generalization of convex functions) and (ii) functions satisfying the Poly"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.10311","kind":"arxiv","version":3},"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/2006.10311/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":"2006.10311","created_at":"2026-07-05T02:25:10.346801+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.10311v3","created_at":"2026-07-05T02:25:10.346801+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.10311","created_at":"2026-07-05T02:25:10.346801+00:00"},{"alias_kind":"pith_short_12","alias_value":"QGCOMYEAW6KY","created_at":"2026-07-05T02:25:10.346801+00:00"},{"alias_kind":"pith_short_16","alias_value":"QGCOMYEAW6KYC5FN","created_at":"2026-07-05T02:25:10.346801+00:00"},{"alias_kind":"pith_short_8","alias_value":"QGCOMYEA","created_at":"2026-07-05T02:25:10.346801+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.08980","citing_title":"Muon Does Not Converge on Convex Lipschitz Functions","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL","json":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL.json","graph_json":"https://pith.science/api/pith-number/QGCOMYEAW6KYC5FNYP57NT23SL/graph.json","events_json":"https://pith.science/api/pith-number/QGCOMYEAW6KYC5FNYP57NT23SL/events.json","paper":"https://pith.science/paper/QGCOMYEA"},"agent_actions":{"view_html":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL","download_json":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL.json","view_paper":"https://pith.science/paper/QGCOMYEA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.10311&json=true","fetch_graph":"https://pith.science/api/pith-number/QGCOMYEAW6KYC5FNYP57NT23SL/graph.json","fetch_events":"https://pith.science/api/pith-number/QGCOMYEAW6KYC5FNYP57NT23SL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL/action/storage_attestation","attest_author":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL/action/author_attestation","sign_citation":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL/action/citation_signature","submit_replication":"https://pith.science/pith/QGCOMYEAW6KYC5FNYP57NT23SL/action/replication_record"}},"created_at":"2026-07-05T02:25:10.346801+00:00","updated_at":"2026-07-05T02:25:10.346801+00:00"}