{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:4EFJ5B5EHTX3IVKCOKFCJJ2NDA","short_pith_number":"pith:4EFJ5B5E","schema_version":"1.0","canonical_sha256":"e10a9e87a43cefb45542728a24a74d18181a2ff69c40c460511449ebcc517d7f","source":{"kind":"arxiv","id":"1907.06246","version":1},"attestation_state":"computed","paper":{"title":"On the Global Convergence of Actor-Critic: A Case for Linear Quadratic Regulator with Ergodic Cost","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Mingyi Hong, Yongxin Chen, Zhaoran Wang, Zhuoran Yang","submitted_at":"2019-07-14T16:50:26Z","abstract_excerpt":"Despite the empirical success of the actor-critic algorithm, its theoretical understanding lags behind. In a broader context, actor-critic can be viewed as an online alternating update algorithm for bilevel optimization, whose convergence is known to be fragile. To understand the instability of actor-critic, we focus on its application to linear quadratic regulators, a simple yet fundamental setting of reinforcement learning. We establish a nonasymptotic convergence analysis of actor-critic in this setting. In particular, we prove that actor-critic finds a globally optimal pair of actor (polic"},"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":"1907.06246","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-14T16:50:26Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"dbd2f15c24f696eca1f93d5c533eaf979e139c67fea5724c2564e2c98ea7b398","abstract_canon_sha256":"de7f5c8e8555c8d85b1dfc3e239fea5282493663d8906f21c7c16a0c700024e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:40:38.133941Z","signature_b64":"/pHpQpaMccyekzfiZ713++2R3wjtDLi896OQSsvJzTweWV0/yvWvxO0MyAe+MzNaaY0pJfcY4ufxvxHDE6q6AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e10a9e87a43cefb45542728a24a74d18181a2ff69c40c460511449ebcc517d7f","last_reissued_at":"2026-05-17T23:40:38.133423Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:40:38.133423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Global Convergence of Actor-Critic: A Case for Linear Quadratic Regulator with Ergodic Cost","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Mingyi Hong, Yongxin Chen, Zhaoran Wang, Zhuoran Yang","submitted_at":"2019-07-14T16:50:26Z","abstract_excerpt":"Despite the empirical success of the actor-critic algorithm, its theoretical understanding lags behind. In a broader context, actor-critic can be viewed as an online alternating update algorithm for bilevel optimization, whose convergence is known to be fragile. To understand the instability of actor-critic, we focus on its application to linear quadratic regulators, a simple yet fundamental setting of reinforcement learning. We establish a nonasymptotic convergence analysis of actor-critic in this setting. In particular, we prove that actor-critic finds a globally optimal pair of actor (polic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.06246","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":"1907.06246","created_at":"2026-05-17T23:40:38.133505+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.06246v1","created_at":"2026-05-17T23:40:38.133505+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.06246","created_at":"2026-05-17T23:40:38.133505+00:00"},{"alias_kind":"pith_short_12","alias_value":"4EFJ5B5EHTX3","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_16","alias_value":"4EFJ5B5EHTX3IVKC","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_8","alias_value":"4EFJ5B5E","created_at":"2026-05-18T12:33:10.108867+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2602.01505","citing_title":"Optimal Sample Complexity for Single Time-Scale Actor-Critic with Momentum","ref_index":61,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA","json":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA.json","graph_json":"https://pith.science/api/pith-number/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/graph.json","events_json":"https://pith.science/api/pith-number/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/events.json","paper":"https://pith.science/paper/4EFJ5B5E"},"agent_actions":{"view_html":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA","download_json":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA.json","view_paper":"https://pith.science/paper/4EFJ5B5E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.06246&json=true","fetch_graph":"https://pith.science/api/pith-number/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/graph.json","fetch_events":"https://pith.science/api/pith-number/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/action/storage_attestation","attest_author":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/action/author_attestation","sign_citation":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/action/citation_signature","submit_replication":"https://pith.science/pith/4EFJ5B5EHTX3IVKCOKFCJJ2NDA/action/replication_record"}},"created_at":"2026-05-17T23:40:38.133505+00:00","updated_at":"2026-05-17T23:40:38.133505+00:00"}