{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:WPUHZ2WSX4COJ6GBYC4ULDPAHZ","short_pith_number":"pith:WPUHZ2WS","schema_version":"1.0","canonical_sha256":"b3e87cead2bf04e4f8c1c0b9458de03e5b423863586037451024ef522acd2ff7","source":{"kind":"arxiv","id":"1802.06480","version":1},"attestation_state":"computed","paper":{"title":"Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Eytan Modiano, Fanyu Que, Qingkai Liang","submitted_at":"2018-02-19T01:25:26Z","abstract_excerpt":"Constrained Markov Decision Process (CMDP) is a natural framework for reinforcement learning tasks with safety constraints, where agents learn a policy that maximizes the long-term reward while satisfying the constraints on the long-term cost. A canonical approach for solving CMDPs is the primal-dual method which updates parameters in primal and dual spaces in turn. Existing methods for CMDPs only use on-policy data for dual updates, which results in sample inefficiency and slow convergence. In this paper, we propose a policy search method for CMDPs called Accelerated Primal-Dual Optimization "},"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":"1802.06480","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-02-19T01:25:26Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"7a95e161639567cfe2260b2229a203ef5b8c12c3c263b81ace0fc7c78eba92d1","abstract_canon_sha256":"c54d1e91a3fa30d8425cf11336a9dad96b927efa3d9b3df7350ef8c781121d0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:23:03.167575Z","signature_b64":"QBG+CClkJ18rbuEPLAqihF7XvduJ+EafQKVc85BIeAXqFjB4L7Xy72Do4YvLX3rCyUZKm6TTp2Z3dR+lf2nDAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b3e87cead2bf04e4f8c1c0b9458de03e5b423863586037451024ef522acd2ff7","last_reissued_at":"2026-05-18T00:23:03.167035Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:23:03.167035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Eytan Modiano, Fanyu Que, Qingkai Liang","submitted_at":"2018-02-19T01:25:26Z","abstract_excerpt":"Constrained Markov Decision Process (CMDP) is a natural framework for reinforcement learning tasks with safety constraints, where agents learn a policy that maximizes the long-term reward while satisfying the constraints on the long-term cost. A canonical approach for solving CMDPs is the primal-dual method which updates parameters in primal and dual spaces in turn. Existing methods for CMDPs only use on-policy data for dual updates, which results in sample inefficiency and slow convergence. In this paper, we propose a policy search method for CMDPs called Accelerated Primal-Dual Optimization "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.06480","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":"1802.06480","created_at":"2026-05-18T00:23:03.167123+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.06480v1","created_at":"2026-05-18T00:23:03.167123+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.06480","created_at":"2026-05-18T00:23:03.167123+00:00"},{"alias_kind":"pith_short_12","alias_value":"WPUHZ2WSX4CO","created_at":"2026-05-18T12:33:01.666342+00:00"},{"alias_kind":"pith_short_16","alias_value":"WPUHZ2WSX4COJ6GB","created_at":"2026-05-18T12:33:01.666342+00:00"},{"alias_kind":"pith_short_8","alias_value":"WPUHZ2WS","created_at":"2026-05-18T12:33:01.666342+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/WPUHZ2WSX4COJ6GBYC4ULDPAHZ","json":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ.json","graph_json":"https://pith.science/api/pith-number/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/graph.json","events_json":"https://pith.science/api/pith-number/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/events.json","paper":"https://pith.science/paper/WPUHZ2WS"},"agent_actions":{"view_html":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ","download_json":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ.json","view_paper":"https://pith.science/paper/WPUHZ2WS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.06480&json=true","fetch_graph":"https://pith.science/api/pith-number/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/graph.json","fetch_events":"https://pith.science/api/pith-number/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/action/storage_attestation","attest_author":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/action/author_attestation","sign_citation":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/action/citation_signature","submit_replication":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/action/replication_record"}},"created_at":"2026-05-18T00:23:03.167123+00:00","updated_at":"2026-05-18T00:23:03.167123+00:00"}