{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:FJF72XD5XDRSEWWEZ3BA2EZAEJ","short_pith_number":"pith:FJF72XD5","schema_version":"1.0","canonical_sha256":"2a4bfd5c7db8e3225ac4cec20d13202259fb17614c07839abcba3c08890fa1e7","source":{"kind":"arxiv","id":"2205.10330","version":5},"attestation_state":"computed","paper":{"title":"A Review of Safe Reinforcement Learning: Methods, Theory and Applications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alois Knoll, Florian Walter, Guang Chen, Jun Wang, Long Yang, Shangding Gu, Yali Du","submitted_at":"2022-05-20T17:42:38Z","abstract_excerpt":"Reinforcement Learning (RL) has achieved tremendous success in many complex decision-making tasks. However, safety concerns are raised during deploying RL in real-world applications, leading to a growing demand for safe RL algorithms, such as in autonomous driving and robotics scenarios. While safe control has a long history, the study of safe RL algorithms is still in the early stages. To establish a good foundation for future safe RL research, in this paper, we provide a review of safe RL from the perspectives of methods, theories, and applications. Firstly, we review the progress of safe RL"},"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":"2205.10330","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2022-05-20T17:42:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"219119a6fa9fbb14d32b77777ff56427299cde0805eeeefc3f787dafd3eeb2be","abstract_canon_sha256":"44e60662697f9c8fadbf52248608f7eba3b4da4472ff27ed25b1553fa24b469b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:54.340077Z","signature_b64":"oGfvb+kWKbqYsilARSdEZyWDr8O6dj/PKg67VtIPRxNFm8fpPqkgtqd50SJzZ1wPnBriuak6Sqr49seHSAs+Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a4bfd5c7db8e3225ac4cec20d13202259fb17614c07839abcba3c08890fa1e7","last_reissued_at":"2026-07-05T08:22:54.339648Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:54.339648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Review of Safe Reinforcement Learning: Methods, Theory and Applications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alois Knoll, Florian Walter, Guang Chen, Jun Wang, Long Yang, Shangding Gu, Yali Du","submitted_at":"2022-05-20T17:42:38Z","abstract_excerpt":"Reinforcement Learning (RL) has achieved tremendous success in many complex decision-making tasks. However, safety concerns are raised during deploying RL in real-world applications, leading to a growing demand for safe RL algorithms, such as in autonomous driving and robotics scenarios. While safe control has a long history, the study of safe RL algorithms is still in the early stages. To establish a good foundation for future safe RL research, in this paper, we provide a review of safe RL from the perspectives of methods, theories, and applications. Firstly, we review the progress of safe RL"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.10330","kind":"arxiv","version":5},"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/2205.10330/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":"2205.10330","created_at":"2026-07-05T08:22:54.339708+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.10330v5","created_at":"2026-07-05T08:22:54.339708+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.10330","created_at":"2026-07-05T08:22:54.339708+00:00"},{"alias_kind":"pith_short_12","alias_value":"FJF72XD5XDRS","created_at":"2026-07-05T08:22:54.339708+00:00"},{"alias_kind":"pith_short_16","alias_value":"FJF72XD5XDRSEWWE","created_at":"2026-07-05T08:22:54.339708+00:00"},{"alias_kind":"pith_short_8","alias_value":"FJF72XD5","created_at":"2026-07-05T08:22:54.339708+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28152","citing_title":"Regularized Reward-Punishment Reinforcement Learning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2503.05724","citing_title":"Addressing Moral Uncertainty using Large Language Models for Ethical Decision-Making","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2503.03480","citing_title":"SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18841","citing_title":"From Cumulative Constraints to Adaptive Runtime Safety Control for Nonstationary Reinforcement Learning","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18842","citing_title":"Safe Continual Reinforcement Learning under Nonstationarity via Adaptive Safety Constraints","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2508.09128","citing_title":"A Review On Safe Reinforcement Learning Using Lyapunov and Barrier Functions","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2509.01728","citing_title":"Constrained Decoding for Safe Robot Navigation Foundation Models","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2510.01020","citing_title":"The Good, the Bad, and the Sampled: a No-Regret Approach to Safe Online Classification","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02727","citing_title":"Data-Driven Synthesis of Probabilistic Controlled Invariant Sets for Linear MDPs","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27861","citing_title":"TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22244","citing_title":"Learning Control Policies to Provably Satisfy Hard Affine Constraints for Black-Box Hybrid Dynamical Systems","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17240","citing_title":"Safe and Policy-Compliant Multi-Agent Orchestration for Enterprise AI","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ","json":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ.json","graph_json":"https://pith.science/api/pith-number/FJF72XD5XDRSEWWEZ3BA2EZAEJ/graph.json","events_json":"https://pith.science/api/pith-number/FJF72XD5XDRSEWWEZ3BA2EZAEJ/events.json","paper":"https://pith.science/paper/FJF72XD5"},"agent_actions":{"view_html":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ","download_json":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ.json","view_paper":"https://pith.science/paper/FJF72XD5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.10330&json=true","fetch_graph":"https://pith.science/api/pith-number/FJF72XD5XDRSEWWEZ3BA2EZAEJ/graph.json","fetch_events":"https://pith.science/api/pith-number/FJF72XD5XDRSEWWEZ3BA2EZAEJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ/action/storage_attestation","attest_author":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ/action/author_attestation","sign_citation":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ/action/citation_signature","submit_replication":"https://pith.science/pith/FJF72XD5XDRSEWWEZ3BA2EZAEJ/action/replication_record"}},"created_at":"2026-07-05T08:22:54.339708+00:00","updated_at":"2026-07-05T08:22:54.339708+00:00"}