{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:GY3C3VXZTFWEPM24EXRUTGEU3I","short_pith_number":"pith:GY3C3VXZ","schema_version":"1.0","canonical_sha256":"36362dd6f9996c47b35c25e3499894da08e749a1170ca2341d94595241519522","source":{"kind":"arxiv","id":"1906.01786","version":3},"attestation_state":"computed","paper":{"title":"Global Optimality Guarantees For Policy Gradient Methods","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Russo, Jalaj Bhandari","submitted_at":"2019-06-05T02:12:22Z","abstract_excerpt":"Policy gradients methods apply to complex, poorly understood, control problems by performing stochastic gradient descent over a parameterized class of polices. Unfortunately, even for simple control problems solvable by standard dynamic programming techniques, policy gradient algorithms face non-convex optimization problems and are widely understood to converge only to a stationary point. This work identifies structural properties -- shared by several classic control problems -- that ensure the policy gradient objective function has no suboptimal stationary points despite being non-convex. Whe"},"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":"1906.01786","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-05T02:12:22Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a42133fc66844df75eeb86a1371eb5b166ab080f8ab7c8f6fe61dd1bc342450c","abstract_canon_sha256":"fc9b650615e16e91c40b2ac22201e6ffb9b1519cec7b51c954507ffcc52a12d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:32:46.235421Z","signature_b64":"dFc8CZusKyXKTWUSryHuzn3pv6aLvLLI70I3iBUDutG4sjVU9a1GvfNzKanQt8kGTf4qi3TG40W0w6gzwr77Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"36362dd6f9996c47b35c25e3499894da08e749a1170ca2341d94595241519522","last_reissued_at":"2026-07-05T04:32:46.234925Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:32:46.234925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Global Optimality Guarantees For Policy Gradient Methods","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Russo, Jalaj Bhandari","submitted_at":"2019-06-05T02:12:22Z","abstract_excerpt":"Policy gradients methods apply to complex, poorly understood, control problems by performing stochastic gradient descent over a parameterized class of polices. Unfortunately, even for simple control problems solvable by standard dynamic programming techniques, policy gradient algorithms face non-convex optimization problems and are widely understood to converge only to a stationary point. This work identifies structural properties -- shared by several classic control problems -- that ensure the policy gradient objective function has no suboptimal stationary points despite being non-convex. Whe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.01786","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/1906.01786/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":"1906.01786","created_at":"2026-07-05T04:32:46.234984+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.01786v3","created_at":"2026-07-05T04:32:46.234984+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.01786","created_at":"2026-07-05T04:32:46.234984+00:00"},{"alias_kind":"pith_short_12","alias_value":"GY3C3VXZTFWE","created_at":"2026-07-05T04:32:46.234984+00:00"},{"alias_kind":"pith_short_16","alias_value":"GY3C3VXZTFWEPM24","created_at":"2026-07-05T04:32:46.234984+00:00"},{"alias_kind":"pith_short_8","alias_value":"GY3C3VXZ","created_at":"2026-07-05T04:32:46.234984+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22579","citing_title":"Stationary Robust Mean-Field Games under Model Mismatches","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I","json":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I.json","graph_json":"https://pith.science/api/pith-number/GY3C3VXZTFWEPM24EXRUTGEU3I/graph.json","events_json":"https://pith.science/api/pith-number/GY3C3VXZTFWEPM24EXRUTGEU3I/events.json","paper":"https://pith.science/paper/GY3C3VXZ"},"agent_actions":{"view_html":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I","download_json":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I.json","view_paper":"https://pith.science/paper/GY3C3VXZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.01786&json=true","fetch_graph":"https://pith.science/api/pith-number/GY3C3VXZTFWEPM24EXRUTGEU3I/graph.json","fetch_events":"https://pith.science/api/pith-number/GY3C3VXZTFWEPM24EXRUTGEU3I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I/action/storage_attestation","attest_author":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I/action/author_attestation","sign_citation":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I/action/citation_signature","submit_replication":"https://pith.science/pith/GY3C3VXZTFWEPM24EXRUTGEU3I/action/replication_record"}},"created_at":"2026-07-05T04:32:46.234984+00:00","updated_at":"2026-07-05T04:32:46.234984+00:00"}