{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:EYE6V3CAZKG6HJIIVKJSNAJ3GL","short_pith_number":"pith:EYE6V3CA","schema_version":"1.0","canonical_sha256":"2609eaec40ca8de3a508aa9326813b32ec186ab5ead260d0931c7c40acde584c","source":{"kind":"arxiv","id":"1805.09416","version":1},"attestation_state":"computed","paper":{"title":"Adaptive Stochastic Gradient Langevin Dynamics: Taming Convergence and Saddle Point Escape Time","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hejian Sang, Jia Liu","submitted_at":"2018-05-23T20:26:56Z","abstract_excerpt":"In this paper, we propose a new adaptive stochastic gradient Langevin dynamics (ASGLD) algorithmic framework and its two specialized versions, namely adaptive stochastic gradient (ASG) and adaptive gradient Langevin dynamics(AGLD), for non-convex optimization problems. All proposed algorithms can escape from saddle points with at most $O(\\log d)$ iterations, which is nearly dimension-free. Further, we show that ASGLD and ASG converge to a local minimum with at most $O(\\log d/\\epsilon^4)$ iterations. Also, ASGLD with full gradients or ASGLD with a slowly linearly increasing batch size converge "},"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":"1805.09416","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2018-05-23T20:26:56Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"765b30d393f3a2fd69c21f44d9f0e475d5c558eb538c525817d3e86aa7139907","abstract_canon_sha256":"35fdc32cf22b23b8bc145b7278ddbff1a17c906be9c6b8f5c7723f9f41cbc0c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:15:04.033961Z","signature_b64":"ekLAFQbcxZ9ATvbrrJRhKOOS9Ur1sQ/8xxN7ceOAzjmDWQrHuph5zO7Vf3K44US25nrwu1a1hnZ30Iu4KTmAAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2609eaec40ca8de3a508aa9326813b32ec186ab5ead260d0931c7c40acde584c","last_reissued_at":"2026-05-18T00:15:04.033327Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:15:04.033327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive Stochastic Gradient Langevin Dynamics: Taming Convergence and Saddle Point Escape Time","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hejian Sang, Jia Liu","submitted_at":"2018-05-23T20:26:56Z","abstract_excerpt":"In this paper, we propose a new adaptive stochastic gradient Langevin dynamics (ASGLD) algorithmic framework and its two specialized versions, namely adaptive stochastic gradient (ASG) and adaptive gradient Langevin dynamics(AGLD), for non-convex optimization problems. All proposed algorithms can escape from saddle points with at most $O(\\log d)$ iterations, which is nearly dimension-free. Further, we show that ASGLD and ASG converge to a local minimum with at most $O(\\log d/\\epsilon^4)$ iterations. Also, ASGLD with full gradients or ASGLD with a slowly linearly increasing batch size converge "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.09416","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":""},"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":"1805.09416","created_at":"2026-05-18T00:15:04.033417+00:00"},{"alias_kind":"arxiv_version","alias_value":"1805.09416v1","created_at":"2026-05-18T00:15:04.033417+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.09416","created_at":"2026-05-18T00:15:04.033417+00:00"},{"alias_kind":"pith_short_12","alias_value":"EYE6V3CAZKG6","created_at":"2026-05-18T12:32:22.470017+00:00"},{"alias_kind":"pith_short_16","alias_value":"EYE6V3CAZKG6HJII","created_at":"2026-05-18T12:32:22.470017+00:00"},{"alias_kind":"pith_short_8","alias_value":"EYE6V3CA","created_at":"2026-05-18T12:32:22.470017+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL","json":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL.json","graph_json":"https://pith.science/api/pith-number/EYE6V3CAZKG6HJIIVKJSNAJ3GL/graph.json","events_json":"https://pith.science/api/pith-number/EYE6V3CAZKG6HJIIVKJSNAJ3GL/events.json","paper":"https://pith.science/paper/EYE6V3CA"},"agent_actions":{"view_html":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL","download_json":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL.json","view_paper":"https://pith.science/paper/EYE6V3CA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1805.09416&json=true","fetch_graph":"https://pith.science/api/pith-number/EYE6V3CAZKG6HJIIVKJSNAJ3GL/graph.json","fetch_events":"https://pith.science/api/pith-number/EYE6V3CAZKG6HJIIVKJSNAJ3GL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL/action/storage_attestation","attest_author":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL/action/author_attestation","sign_citation":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL/action/citation_signature","submit_replication":"https://pith.science/pith/EYE6V3CAZKG6HJIIVKJSNAJ3GL/action/replication_record"}},"created_at":"2026-05-18T00:15:04.033417+00:00","updated_at":"2026-05-18T00:15:04.033417+00:00"}