{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:ZB7WSCJDVRCXDEAP6OO3OGPT2I","short_pith_number":"pith:ZB7WSCJD","schema_version":"1.0","canonical_sha256":"c87f690923ac4571900ff39db719f3d22c5650cbd88628cfc6d4165dca58be27","source":{"kind":"arxiv","id":"1707.06618","version":3},"attestation_state":"computed","paper":{"title":"Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Difan Zou, Jinghui Chen, Pan Xu, Quanquan Gu","submitted_at":"2017-07-20T17:18:11Z","abstract_excerpt":"We present a unified framework to analyze the global convergence of Langevin dynamics based algorithms for nonconvex finite-sum optimization with $n$ component functions. At the core of our analysis is a direct analysis of the ergodicity of the numerical approximations to Langevin dynamics, which leads to faster convergence rates. Specifically, we show that gradient Langevin dynamics (GLD) and stochastic gradient Langevin dynamics (SGLD) converge to the almost minimizer within $\\tilde O\\big(nd/(\\lambda\\epsilon) \\big)$ and $\\tilde O\\big(d^7/(\\lambda^5\\epsilon^5) \\big)$ stochastic gradient evalu"},"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":"1707.06618","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-07-20T17:18:11Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"40ed6647800900d536ad24728dc05ef162260271d91b419992837c1f746a658d","abstract_canon_sha256":"acff87a147f4badf33bcbd3f74a5cc81d036c2227517ac8def74c4f255cdb5b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:43:53.487291Z","signature_b64":"gJihYmwOc7g4QHSbYcO6r9VlmFWbq7RWkT9zjs5CxpTNqF5ORIkBePHKTf5PLpqPAk3zMQCJAx61Xjxr8qIWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c87f690923ac4571900ff39db719f3d22c5650cbd88628cfc6d4165dca58be27","last_reissued_at":"2026-07-05T01:43:53.486852Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:43:53.486852Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Difan Zou, Jinghui Chen, Pan Xu, Quanquan Gu","submitted_at":"2017-07-20T17:18:11Z","abstract_excerpt":"We present a unified framework to analyze the global convergence of Langevin dynamics based algorithms for nonconvex finite-sum optimization with $n$ component functions. At the core of our analysis is a direct analysis of the ergodicity of the numerical approximations to Langevin dynamics, which leads to faster convergence rates. Specifically, we show that gradient Langevin dynamics (GLD) and stochastic gradient Langevin dynamics (SGLD) converge to the almost minimizer within $\\tilde O\\big(nd/(\\lambda\\epsilon) \\big)$ and $\\tilde O\\big(d^7/(\\lambda^5\\epsilon^5) \\big)$ stochastic gradient evalu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1707.06618","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/1707.06618/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":"1707.06618","created_at":"2026-07-05T01:43:53.486912+00:00"},{"alias_kind":"arxiv_version","alias_value":"1707.06618v3","created_at":"2026-07-05T01:43:53.486912+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1707.06618","created_at":"2026-07-05T01:43:53.486912+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZB7WSCJDVRCX","created_at":"2026-07-05T01:43:53.486912+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZB7WSCJDVRCXDEAP","created_at":"2026-07-05T01:43:53.486912+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZB7WSCJD","created_at":"2026-07-05T01:43:53.486912+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2110.12922","citing_title":"On quantitative Laplace-type convergence results for some exponential probability measures, with two applications","ref_index":72,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I","json":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I.json","graph_json":"https://pith.science/api/pith-number/ZB7WSCJDVRCXDEAP6OO3OGPT2I/graph.json","events_json":"https://pith.science/api/pith-number/ZB7WSCJDVRCXDEAP6OO3OGPT2I/events.json","paper":"https://pith.science/paper/ZB7WSCJD"},"agent_actions":{"view_html":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I","download_json":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I.json","view_paper":"https://pith.science/paper/ZB7WSCJD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1707.06618&json=true","fetch_graph":"https://pith.science/api/pith-number/ZB7WSCJDVRCXDEAP6OO3OGPT2I/graph.json","fetch_events":"https://pith.science/api/pith-number/ZB7WSCJDVRCXDEAP6OO3OGPT2I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I/action/storage_attestation","attest_author":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I/action/author_attestation","sign_citation":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I/action/citation_signature","submit_replication":"https://pith.science/pith/ZB7WSCJDVRCXDEAP6OO3OGPT2I/action/replication_record"}},"created_at":"2026-07-05T01:43:53.486912+00:00","updated_at":"2026-07-05T01:43:53.486912+00:00"}