{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:KHRBY5QFSYSFXF247YM3C3ZPHB","short_pith_number":"pith:KHRBY5QF","schema_version":"1.0","canonical_sha256":"51e21c760596245b975cfe19b16f2f385c3826eb237793d17bc947403cd86a75","source":{"kind":"arxiv","id":"1911.09101","version":2},"attestation_state":"computed","paper":{"title":"Safe Policies for Reinforcement Learning via Primal-Dual Methods","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SY","math.OC"],"primary_cat":"eess.SY","authors_text":"Alejandro Ribeiro, Luiz F. O. Chamon, Miguel Calvo-Fullana, Santiago Paternain","submitted_at":"2019-11-20T17:56:39Z","abstract_excerpt":"In this paper, we study the learning of safe policies in the setting of reinforcement learning problems. This is, we aim to control a Markov Decision Process (MDP) of which we do not know the transition probabilities, but we have access to sample trajectories through experience. We define safety as the agent remaining in a desired safe set with high probability during the operation time. We therefore consider a constrained MDP where the constraints are probabilistic. Since there is no straightforward way to optimize the policy with respect to the probabilistic constraint in a reinforcement lea"},"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":"1911.09101","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SY","submitted_at":"2019-11-20T17:56:39Z","cross_cats_sorted":["cs.LG","cs.SY","math.OC"],"title_canon_sha256":"8a6d2ebc798decc2aa8b6b3343720476344f736479e4be06826d39e1233227e2","abstract_canon_sha256":"4f1edf88bffbb6b73558b4b251d65fc46d3e92a74334edfe71a006dda284d15d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:48:03.306774Z","signature_b64":"ypxK5uEqIHKs21k3/c/LEWXzXFpPm8tY6GjGRixOlyI1BioWtJIuH8p46wfbri5kaY/MqyEvRNOx/DoyVnFmBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51e21c760596245b975cfe19b16f2f385c3826eb237793d17bc947403cd86a75","last_reissued_at":"2026-07-05T03:48:03.306194Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:48:03.306194Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Safe Policies for Reinforcement Learning via Primal-Dual Methods","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SY","math.OC"],"primary_cat":"eess.SY","authors_text":"Alejandro Ribeiro, Luiz F. O. Chamon, Miguel Calvo-Fullana, Santiago Paternain","submitted_at":"2019-11-20T17:56:39Z","abstract_excerpt":"In this paper, we study the learning of safe policies in the setting of reinforcement learning problems. This is, we aim to control a Markov Decision Process (MDP) of which we do not know the transition probabilities, but we have access to sample trajectories through experience. We define safety as the agent remaining in a desired safe set with high probability during the operation time. We therefore consider a constrained MDP where the constraints are probabilistic. Since there is no straightforward way to optimize the policy with respect to the probabilistic constraint in a reinforcement lea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.09101","kind":"arxiv","version":2},"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/1911.09101/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":"1911.09101","created_at":"2026-07-05T03:48:03.306253+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.09101v2","created_at":"2026-07-05T03:48:03.306253+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.09101","created_at":"2026-07-05T03:48:03.306253+00:00"},{"alias_kind":"pith_short_12","alias_value":"KHRBY5QFSYSF","created_at":"2026-07-05T03:48:03.306253+00:00"},{"alias_kind":"pith_short_16","alias_value":"KHRBY5QFSYSFXF24","created_at":"2026-07-05T03:48:03.306253+00:00"},{"alias_kind":"pith_short_8","alias_value":"KHRBY5QF","created_at":"2026-07-05T03:48:03.306253+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.21852","citing_title":"A Provable Approach for End-to-End Safe Reinforcement Learning","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB","json":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB.json","graph_json":"https://pith.science/api/pith-number/KHRBY5QFSYSFXF247YM3C3ZPHB/graph.json","events_json":"https://pith.science/api/pith-number/KHRBY5QFSYSFXF247YM3C3ZPHB/events.json","paper":"https://pith.science/paper/KHRBY5QF"},"agent_actions":{"view_html":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB","download_json":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB.json","view_paper":"https://pith.science/paper/KHRBY5QF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.09101&json=true","fetch_graph":"https://pith.science/api/pith-number/KHRBY5QFSYSFXF247YM3C3ZPHB/graph.json","fetch_events":"https://pith.science/api/pith-number/KHRBY5QFSYSFXF247YM3C3ZPHB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB/action/storage_attestation","attest_author":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB/action/author_attestation","sign_citation":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB/action/citation_signature","submit_replication":"https://pith.science/pith/KHRBY5QFSYSFXF247YM3C3ZPHB/action/replication_record"}},"created_at":"2026-07-05T03:48:03.306253+00:00","updated_at":"2026-07-05T03:48:03.306253+00:00"}