{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:WPUHZ2WSX4COJ6GBYC4ULDPAHZ","short_pith_number":"pith:WPUHZ2WS","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"},"canonical_sha256":"b3e87cead2bf04e4f8c1c0b9458de03e5b423863586037451024ef522acd2ff7","source":{"kind":"arxiv","id":"1802.06480","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1802.06480","created_at":"2026-05-18T00:23:03Z"},{"alias_kind":"arxiv_version","alias_value":"1802.06480v1","created_at":"2026-05-18T00:23:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.06480","created_at":"2026-05-18T00:23:03Z"},{"alias_kind":"pith_short_12","alias_value":"WPUHZ2WSX4CO","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"WPUHZ2WSX4COJ6GB","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"WPUHZ2WS","created_at":"2026-05-18T12:33:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:WPUHZ2WSX4COJ6GBYC4ULDPAHZ","target":"record","payload":{"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"},"canonical_sha256":"b3e87cead2bf04e4f8c1c0b9458de03e5b423863586037451024ef522acd2ff7","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"},"source_kind":"arxiv","source_id":"1802.06480","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:23:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qAzd18YmfTY9rUaquyTTzJQ5eQ6SX0OQmRAGjoVpij9s/IvZbFMFuV5rwnsRMdf/cpt9O0H3gsRPpaZGKvL2BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T15:23:31.586538Z"},"content_sha256":"519aac4b654464290dc6d1fba12991a7456ed5b32d66fc9e3dcb8449f4cf8940","schema_version":"1.0","event_id":"sha256:519aac4b654464290dc6d1fba12991a7456ed5b32d66fc9e3dcb8449f4cf8940"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:WPUHZ2WSX4COJ6GBYC4ULDPAHZ","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:23:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IgTaNNi/OsFW1um0RgiqBn584ZmBF0gHOXbTSFfMtnJc5iXJWbMPCdDuos9SedhPi8Frr/tqansjuJM71+KzDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T15:23:31.586894Z"},"content_sha256":"4de9e252a1c4cb0865d0cff7f51422a99f042a6e1a0b8dd92162d4cb85b7add6","schema_version":"1.0","event_id":"sha256:4de9e252a1c4cb0865d0cff7f51422a99f042a6e1a0b8dd92162d4cb85b7add6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/bundle.json","state_url":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-27T15:23:31Z","links":{"resolver":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ","bundle":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/bundle.json","state":"https://pith.science/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WPUHZ2WSX4COJ6GBYC4ULDPAHZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:WPUHZ2WSX4COJ6GBYC4ULDPAHZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c54d1e91a3fa30d8425cf11336a9dad96b927efa3d9b3df7350ef8c781121d0d","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-02-19T01:25:26Z","title_canon_sha256":"7a95e161639567cfe2260b2229a203ef5b8c12c3c263b81ace0fc7c78eba92d1"},"schema_version":"1.0","source":{"id":"1802.06480","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1802.06480","created_at":"2026-05-18T00:23:03Z"},{"alias_kind":"arxiv_version","alias_value":"1802.06480v1","created_at":"2026-05-18T00:23:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.06480","created_at":"2026-05-18T00:23:03Z"},{"alias_kind":"pith_short_12","alias_value":"WPUHZ2WSX4CO","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"WPUHZ2WSX4COJ6GB","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"WPUHZ2WS","created_at":"2026-05-18T12:33:01Z"}],"graph_snapshots":[{"event_id":"sha256:4de9e252a1c4cb0865d0cff7f51422a99f042a6e1a0b8dd92162d4cb85b7add6","target":"graph","created_at":"2026-05-18T00:23:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"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 ","authors_text":"Eytan Modiano, Fanyu Que, Qingkai Liang","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-02-19T01:25:26Z","title":"Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.06480","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:519aac4b654464290dc6d1fba12991a7456ed5b32d66fc9e3dcb8449f4cf8940","target":"record","created_at":"2026-05-18T00:23:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c54d1e91a3fa30d8425cf11336a9dad96b927efa3d9b3df7350ef8c781121d0d","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-02-19T01:25:26Z","title_canon_sha256":"7a95e161639567cfe2260b2229a203ef5b8c12c3c263b81ace0fc7c78eba92d1"},"schema_version":"1.0","source":{"id":"1802.06480","kind":"arxiv","version":1}},"canonical_sha256":"b3e87cead2bf04e4f8c1c0b9458de03e5b423863586037451024ef522acd2ff7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b3e87cead2bf04e4f8c1c0b9458de03e5b423863586037451024ef522acd2ff7","first_computed_at":"2026-05-18T00:23:03.167035Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:23:03.167035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QBG+CClkJ18rbuEPLAqihF7XvduJ+EafQKVc85BIeAXqFjB4L7Xy72Do4YvLX3rCyUZKm6TTp2Z3dR+lf2nDAQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:23:03.167575Z","signed_message":"canonical_sha256_bytes"},"source_id":"1802.06480","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:519aac4b654464290dc6d1fba12991a7456ed5b32d66fc9e3dcb8449f4cf8940","sha256:4de9e252a1c4cb0865d0cff7f51422a99f042a6e1a0b8dd92162d4cb85b7add6"],"state_sha256":"aafe4380f13e50505c43cc60f834bc059f31724b349f9ed96641654e9b4d3a7d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F5OJeFCWkVYvh27LLzOCZ/xIj6PcUeV+z4A/TSUHLHuLJMNVRT37/o3eD/KcLRiQtGI94WKspd3dlN0DWVljDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T15:23:31.590533Z","bundle_sha256":"06195eecab18ab4d7bc2ef392662e1f6ed28538b776e7a880fc594487b1beec5"}}