{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4B5OHG54IYYTZYLTF4M7IE2L3I","short_pith_number":"pith:4B5OHG54","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"},"canonical_sha256":"e07ae39bbc46313ce1732f19f4134bda145ae95f289df91226b39709d93d2157","source":{"kind":"arxiv","id":"2505.01619","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01619","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01619v1","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01619","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_12","alias_value":"4B5OHG54IYYT","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_16","alias_value":"4B5OHG54IYYTZYLT","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_8","alias_value":"4B5OHG54","created_at":"2026-07-05T10:58:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4B5OHG54IYYTZYLTF4M7IE2L3I","target":"record","payload":{"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"},"canonical_sha256":"e07ae39bbc46313ce1732f19f4134bda145ae95f289df91226b39709d93d2157","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"},"source_kind":"arxiv","source_id":"2505.01619","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":"2026-07-05T10:58:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L62on1QWpuJzmBE/4K39+ldqi7AME0XpkCw2lmxCujvMsTGeFrrM7faaOgf1NBbSdvQPXny09PHmVu/iIb0NBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T16:13:58.163354Z"},"content_sha256":"355f209fedf4ca4a6dfc708afceb132cf0d666ba7242a57d1f48ae6258a68ba0","schema_version":"1.0","event_id":"sha256:355f209fedf4ca4a6dfc708afceb132cf0d666ba7242a57d1f48ae6258a68ba0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4B5OHG54IYYTZYLTF4M7IE2L3I","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:58:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7E6qctVy5OBQMkSsjMVM+3xtdlbMtyIG5j7jUcq4OLxR+ASeBwZGZ6OM9IFMs4I4mlVn3Ib0RHlShZkN1yTRDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T16:13:58.163865Z"},"content_sha256":"3e7bfadeec9ff06ea975e9df46f35d5fc5748488da2c5730f73c352b5f89eeff","schema_version":"1.0","event_id":"sha256:3e7bfadeec9ff06ea975e9df46f35d5fc5748488da2c5730f73c352b5f89eeff"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/bundle.json","state_url":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/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":"2026-08-23T16:13:58Z","links":{"resolver":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I","bundle":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/bundle.json","state":"https://pith.science/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4B5OHG54IYYTZYLTF4M7IE2L3I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4B5OHG54IYYTZYLTF4M7IE2L3I","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":"5cc962b6805fe4b4584fbd51be5fb6bad936ba4d23d6a51a090ae9fb3a0a15ac","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T22:48:27Z","title_canon_sha256":"5ff063ab63f6431aa37b7218cd527e813419441e355aa4e2ad67898c12e6c225"},"schema_version":"1.0","source":{"id":"2505.01619","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01619","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01619v1","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01619","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_12","alias_value":"4B5OHG54IYYT","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_16","alias_value":"4B5OHG54IYYTZYLT","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_8","alias_value":"4B5OHG54","created_at":"2026-07-05T10:58:00Z"}],"graph_snapshots":[{"event_id":"sha256:3e7bfadeec9ff06ea975e9df46f35d5fc5748488da2c5730f73c352b5f89eeff","target":"graph","created_at":"2026-07-05T10:58:00Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.01619/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Hanping Zhang, Yuhong Guo","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T22:48:27Z","title":"Skill-based Safe Reinforcement Learning with Risk Planning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01619","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":"sha256:355f209fedf4ca4a6dfc708afceb132cf0d666ba7242a57d1f48ae6258a68ba0","target":"record","created_at":"2026-07-05T10:58:00Z","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":"5cc962b6805fe4b4584fbd51be5fb6bad936ba4d23d6a51a090ae9fb3a0a15ac","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T22:48:27Z","title_canon_sha256":"5ff063ab63f6431aa37b7218cd527e813419441e355aa4e2ad67898c12e6c225"},"schema_version":"1.0","source":{"id":"2505.01619","kind":"arxiv","version":1}},"canonical_sha256":"e07ae39bbc46313ce1732f19f4134bda145ae95f289df91226b39709d93d2157","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e07ae39bbc46313ce1732f19f4134bda145ae95f289df91226b39709d93d2157","first_computed_at":"2026-07-05T10:58:00.857643Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:58:00.857643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GJ9Xq6bafg5Vd2INxCtHOynpkFlGQ6rqUNx2Wm6dqCH7cbxteFoX0rHWmQEM7uLC7ZTe0seKyAKA7bWR92onBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:58:00.858128Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.01619","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:355f209fedf4ca4a6dfc708afceb132cf0d666ba7242a57d1f48ae6258a68ba0","sha256:3e7bfadeec9ff06ea975e9df46f35d5fc5748488da2c5730f73c352b5f89eeff"],"state_sha256":"fd18f310a4e3e67c9c0c9b34813dc6105d1d4ea672cff3fb835372ee683ec688"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wzRFi/SO3m47M2TEHPk8jCNmw5FDk+6yO43afTe45HMJm+OLDvqNVRsMercPuepVdi4vCRVG4bGp78lOPXNHDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T16:13:58.167672Z","bundle_sha256":"8b7ca99cb4bd9d5f2db6f5e64a0a6e27e915a39148ffdaca7561edc7cdcd3f8d"}}