{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4B5OHG54IYYTZYLTF4M7IE2L3I","short_pith_number":"pith:4B5OHG54","schema_version":"1.0","canonical_sha256":"e07ae39bbc46313ce1732f19f4134bda145ae95f289df91226b39709d93d2157","source":{"kind":"arxiv","id":"2505.01619","version":1},"attestation_state":"computed","paper":{"title":"Skill-based Safe Reinforcement Learning with Risk Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hanping Zhang, Yuhong Guo","submitted_at":"2025-05-02T22:48:27Z","abstract_excerpt":"Safe Reinforcement Learning (Safe RL) aims to ensure safety when an RL agent conducts learning by interacting with real-world environments where improper actions can induce high costs or lead to severe consequences. In this paper, we propose a novel Safe Skill Planning (SSkP) approach to enhance effective safe RL by exploiting auxiliary offline demonstration data. SSkP involves a two-stage process. First, we employ PU learning to learn a skill risk predictor from the offline demonstration data. Then, based on the learned skill risk predictor, we develop a novel risk planning process to enhance"},"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":"2505.01619","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T22:48:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5ff063ab63f6431aa37b7218cd527e813419441e355aa4e2ad67898c12e6c225","abstract_canon_sha256":"5cc962b6805fe4b4584fbd51be5fb6bad936ba4d23d6a51a090ae9fb3a0a15ac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:00.858128Z","signature_b64":"GJ9Xq6bafg5Vd2INxCtHOynpkFlGQ6rqUNx2Wm6dqCH7cbxteFoX0rHWmQEM7uLC7ZTe0seKyAKA7bWR92onBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e07ae39bbc46313ce1732f19f4134bda145ae95f289df91226b39709d93d2157","last_reissued_at":"2026-07-05T10:58:00.857643Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:00.857643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Skill-based Safe Reinforcement Learning with Risk Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hanping Zhang, Yuhong Guo","submitted_at":"2025-05-02T22:48:27Z","abstract_excerpt":"Safe Reinforcement Learning (Safe RL) aims to ensure safety when an RL agent conducts learning by interacting with real-world environments where improper actions can induce high costs or lead to severe consequences. In this paper, we propose a novel Safe Skill Planning (SSkP) approach to enhance effective safe RL by exploiting auxiliary offline demonstration data. SSkP involves a two-stage process. First, we employ PU learning to learn a skill risk predictor from the offline demonstration data. Then, based on the learned skill risk predictor, we develop a novel risk planning process to enhance"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01619","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/2505.01619/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":"2505.01619","created_at":"2026-07-05T10:58:00.857703+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.01619v1","created_at":"2026-07-05T10:58:00.857703+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01619","created_at":"2026-07-05T10:58:00.857703+00:00"},{"alias_kind":"pith_short_12","alias_value":"4B5OHG54IYYT","created_at":"2026-07-05T10:58:00.857703+00:00"},{"alias_kind":"pith_short_16","alias_value":"4B5OHG54IYYTZYLT","created_at":"2026-07-05T10:58:00.857703+00:00"},{"alias_kind":"pith_short_8","alias_value":"4B5OHG54","created_at":"2026-07-05T10:58:00.857703+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/4B5OHG54IYYTZYLTF4M7IE2L3I","json":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I.json","graph_json":"https://pith.science/api/pith-number/4B5OHG54IYYTZYLTF4M7IE2L3I/graph.json","events_json":"https://pith.science/api/pith-number/4B5OHG54IYYTZYLTF4M7IE2L3I/events.json","paper":"https://pith.science/paper/4B5OHG54"},"agent_actions":{"view_html":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I","download_json":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I.json","view_paper":"https://pith.science/paper/4B5OHG54","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.01619&json=true","fetch_graph":"https://pith.science/api/pith-number/4B5OHG54IYYTZYLTF4M7IE2L3I/graph.json","fetch_events":"https://pith.science/api/pith-number/4B5OHG54IYYTZYLTF4M7IE2L3I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/action/storage_attestation","attest_author":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/action/author_attestation","sign_citation":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/action/citation_signature","submit_replication":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/action/replication_record"}},"created_at":"2026-07-05T10:58:00.857703+00:00","updated_at":"2026-07-05T10:58:00.857703+00:00"}