{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LOPRAVHX4XHWPHGMTRTSCBEQVG","short_pith_number":"pith:LOPRAVHX","schema_version":"1.0","canonical_sha256":"5b9f1054f7e5cf679ccc9c67210490a98cf7d8c572756e8d1664cb7bf93df85c","source":{"kind":"arxiv","id":"2402.15197","version":1},"attestation_state":"computed","paper":{"title":"Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.RO","cs.SY"],"primary_cat":"eess.SY","authors_text":"Homayoun Honari, Homayoun Najjaran, Mehran Ghafarian Tamizi","submitted_at":"2024-02-23T08:58:38Z","abstract_excerpt":"Safe reinforcement learning (Safe RL) refers to a class of techniques that aim to prevent RL algorithms from violating constraints in the process of decision-making and exploration during trial and error. In this paper, a novel model-free Safe RL algorithm, formulated based on the multi-objective policy optimization framework is introduced where the policy is optimized towards optimality and safety, simultaneously. The optimality is achieved by the environment reward function that is subsequently shaped using a safety critic. The advantage of the Safety Optimized RL (SORL) algorithm compared t"},"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":"2402.15197","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.SY","submitted_at":"2024-02-23T08:58:38Z","cross_cats_sorted":["cs.AI","cs.LG","cs.RO","cs.SY"],"title_canon_sha256":"7191e73daf4a0abc1069b043f521c89f59ff2d08995d97882f015baa79b4254e","abstract_canon_sha256":"6b9c8c908209005f4e17fa16973174fa0aa6bf289764b4030f158d2163900dab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:54:47.633804Z","signature_b64":"mcmjoGWn1qYKY+dIgaxHRwMzDwQUzvYaJ69hEKggCJ8DtLdmUsCkdFzu0ekHAbXXkKOjKBdLFtzv1ousm1NtAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b9f1054f7e5cf679ccc9c67210490a98cf7d8c572756e8d1664cb7bf93df85c","last_reissued_at":"2026-07-05T08:54:47.633247Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:54:47.633247Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.RO","cs.SY"],"primary_cat":"eess.SY","authors_text":"Homayoun Honari, Homayoun Najjaran, Mehran Ghafarian Tamizi","submitted_at":"2024-02-23T08:58:38Z","abstract_excerpt":"Safe reinforcement learning (Safe RL) refers to a class of techniques that aim to prevent RL algorithms from violating constraints in the process of decision-making and exploration during trial and error. In this paper, a novel model-free Safe RL algorithm, formulated based on the multi-objective policy optimization framework is introduced where the policy is optimized towards optimality and safety, simultaneously. The optimality is achieved by the environment reward function that is subsequently shaped using a safety critic. The advantage of the Safety Optimized RL (SORL) algorithm compared t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.15197","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2402.15197/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":"2402.15197","created_at":"2026-07-05T08:54:47.633324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.15197v1","created_at":"2026-07-05T08:54:47.633324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.15197","created_at":"2026-07-05T08:54:47.633324+00:00"},{"alias_kind":"pith_short_12","alias_value":"LOPRAVHX4XHW","created_at":"2026-07-05T08:54:47.633324+00:00"},{"alias_kind":"pith_short_16","alias_value":"LOPRAVHX4XHWPHGM","created_at":"2026-07-05T08:54:47.633324+00:00"},{"alias_kind":"pith_short_8","alias_value":"LOPRAVHX","created_at":"2026-07-05T08:54:47.633324+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.19809","citing_title":"Q-learning-based Model-free Safety Filter","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG","json":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG.json","graph_json":"https://pith.science/api/pith-number/LOPRAVHX4XHWPHGMTRTSCBEQVG/graph.json","events_json":"https://pith.science/api/pith-number/LOPRAVHX4XHWPHGMTRTSCBEQVG/events.json","paper":"https://pith.science/paper/LOPRAVHX"},"agent_actions":{"view_html":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG","download_json":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG.json","view_paper":"https://pith.science/paper/LOPRAVHX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.15197&json=true","fetch_graph":"https://pith.science/api/pith-number/LOPRAVHX4XHWPHGMTRTSCBEQVG/graph.json","fetch_events":"https://pith.science/api/pith-number/LOPRAVHX4XHWPHGMTRTSCBEQVG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG/action/storage_attestation","attest_author":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG/action/author_attestation","sign_citation":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG/action/citation_signature","submit_replication":"https://pith.science/pith/LOPRAVHX4XHWPHGMTRTSCBEQVG/action/replication_record"}},"created_at":"2026-07-05T08:54:47.633324+00:00","updated_at":"2026-07-05T08:54:47.633324+00:00"}