{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:GB3QPWUFKFXB3GNZNUS75EWV4Z","short_pith_number":"pith:GB3QPWUF","canonical_record":{"source":{"id":"2303.03226","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-03-06T15:43:41Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b5c0f828810973fe00a82e44cf2f5841783e213def2d5b2aabaf9274ddf90adb","abstract_canon_sha256":"f76eb270876be8e53e4cd4804bac27d31737ad58c76f495052ea44ff6cf2a868"},"schema_version":"1.0"},"canonical_sha256":"307707da85516e1d99b96d25fe92d5e6519050f81f7821125873b7bd83ce8379","source":{"kind":"arxiv","id":"2303.03226","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.03226","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"arxiv_version","alias_value":"2303.03226v1","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.03226","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"pith_short_12","alias_value":"GB3QPWUFKFXB","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"pith_short_16","alias_value":"GB3QPWUFKFXB3GNZ","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"pith_short_8","alias_value":"GB3QPWUF","created_at":"2026-07-05T05:48:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:GB3QPWUFKFXB3GNZNUS75EWV4Z","target":"record","payload":{"canonical_record":{"source":{"id":"2303.03226","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-03-06T15:43:41Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b5c0f828810973fe00a82e44cf2f5841783e213def2d5b2aabaf9274ddf90adb","abstract_canon_sha256":"f76eb270876be8e53e4cd4804bac27d31737ad58c76f495052ea44ff6cf2a868"},"schema_version":"1.0"},"canonical_sha256":"307707da85516e1d99b96d25fe92d5e6519050f81f7821125873b7bd83ce8379","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:27.198997Z","signature_b64":"6aI2QnBMby1juBIS84iqnjZEGzuawF1eNwlLtn8fX5tgNLq8kYUpyht0RH2v3shTenpwLLbNFqU2hnvz0zJ4BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"307707da85516e1d99b96d25fe92d5e6519050f81f7821125873b7bd83ce8379","last_reissued_at":"2026-07-05T05:48:27.198644Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:27.198644Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.03226","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":"2026-07-05T05:48:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Wyf4pfZLjgHXTWwsFC2cQZaUxaXxHEECTdFuM99WRr4SQ3M1+NIGrj4uSx5mkdrq/IibNdeImHqyRI2I5bacBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T17:50:01.804170Z"},"content_sha256":"9549438f7e6c9c400ca183dd030c38d2f0a3f67709a09c343497cb64708d47ec","schema_version":"1.0","event_id":"sha256:9549438f7e6c9c400ca183dd030c38d2f0a3f67709a09c343497cb64708d47ec"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:GB3QPWUFKFXB3GNZNUS75EWV4Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Safe Reinforcement Learning via Probabilistic Logic Shields","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Gavin Rens, Giuseppe Marra, Luc De Raedt, Wen-Chi Yang","submitted_at":"2023-03-06T15:43:41Z","abstract_excerpt":"Safe Reinforcement learning (Safe RL) aims at learning optimal policies while staying safe. A popular solution to Safe RL is shielding, which uses a logical safety specification to prevent an RL agent from taking unsafe actions. However, traditional shielding techniques are difficult to integrate with continuous, end-to-end deep RL methods. To this end, we introduce Probabilistic Logic Policy Gradient (PLPG). PLPG is a model-based Safe RL technique that uses probabilistic logic programming to model logical safety constraints as differentiable functions. Therefore, PLPG can be seamlessly applie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.03226","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/2303.03226/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":"2026-07-05T05:48:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"veXCFezO05veMJTxxdtM0Bt0m8yAh5YR9g+u7pyGGBfub/PpqG3fcqfV+hacgvBVjWjxjqA+i2jH3ivFeEgYCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T17:50:01.804702Z"},"content_sha256":"55f73f169abe24a70bf8143a665ac1101ff578470bf9efca9b39413e4b0e9d40","schema_version":"1.0","event_id":"sha256:55f73f169abe24a70bf8143a665ac1101ff578470bf9efca9b39413e4b0e9d40"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z/bundle.json","state_url":"https://pith.science/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z/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":"2026-08-19T17:50:01Z","links":{"resolver":"https://pith.science/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z","bundle":"https://pith.science/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z/bundle.json","state":"https://pith.science/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GB3QPWUFKFXB3GNZNUS75EWV4Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:GB3QPWUFKFXB3GNZNUS75EWV4Z","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":"f76eb270876be8e53e4cd4804bac27d31737ad58c76f495052ea44ff6cf2a868","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-03-06T15:43:41Z","title_canon_sha256":"b5c0f828810973fe00a82e44cf2f5841783e213def2d5b2aabaf9274ddf90adb"},"schema_version":"1.0","source":{"id":"2303.03226","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.03226","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"arxiv_version","alias_value":"2303.03226v1","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.03226","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"pith_short_12","alias_value":"GB3QPWUFKFXB","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"pith_short_16","alias_value":"GB3QPWUFKFXB3GNZ","created_at":"2026-07-05T05:48:27Z"},{"alias_kind":"pith_short_8","alias_value":"GB3QPWUF","created_at":"2026-07-05T05:48:27Z"}],"graph_snapshots":[{"event_id":"sha256:55f73f169abe24a70bf8143a665ac1101ff578470bf9efca9b39413e4b0e9d40","target":"graph","created_at":"2026-07-05T05:48:27Z","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/2303.03226/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Safe Reinforcement learning (Safe RL) aims at learning optimal policies while staying safe. A popular solution to Safe RL is shielding, which uses a logical safety specification to prevent an RL agent from taking unsafe actions. However, traditional shielding techniques are difficult to integrate with continuous, end-to-end deep RL methods. To this end, we introduce Probabilistic Logic Policy Gradient (PLPG). PLPG is a model-based Safe RL technique that uses probabilistic logic programming to model logical safety constraints as differentiable functions. Therefore, PLPG can be seamlessly applie","authors_text":"Gavin Rens, Giuseppe Marra, Luc De Raedt, Wen-Chi Yang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-03-06T15:43:41Z","title":"Safe Reinforcement Learning via Probabilistic Logic Shields"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.03226","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":"sha256:9549438f7e6c9c400ca183dd030c38d2f0a3f67709a09c343497cb64708d47ec","target":"record","created_at":"2026-07-05T05:48:27Z","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":"f76eb270876be8e53e4cd4804bac27d31737ad58c76f495052ea44ff6cf2a868","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-03-06T15:43:41Z","title_canon_sha256":"b5c0f828810973fe00a82e44cf2f5841783e213def2d5b2aabaf9274ddf90adb"},"schema_version":"1.0","source":{"id":"2303.03226","kind":"arxiv","version":1}},"canonical_sha256":"307707da85516e1d99b96d25fe92d5e6519050f81f7821125873b7bd83ce8379","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"307707da85516e1d99b96d25fe92d5e6519050f81f7821125873b7bd83ce8379","first_computed_at":"2026-07-05T05:48:27.198644Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:48:27.198644Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6aI2QnBMby1juBIS84iqnjZEGzuawF1eNwlLtn8fX5tgNLq8kYUpyht0RH2v3shTenpwLLbNFqU2hnvz0zJ4BA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:48:27.198997Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.03226","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9549438f7e6c9c400ca183dd030c38d2f0a3f67709a09c343497cb64708d47ec","sha256:55f73f169abe24a70bf8143a665ac1101ff578470bf9efca9b39413e4b0e9d40"],"state_sha256":"c7e8ce7b078df336b7fe418e4805449e84424fc01cc1f051e027ee1a95025786"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"czLXdIb/YipwE/xHes0ef0IjNdIzfFH9vs/HGqXFSqaRpfbzSA0wXTX5z2uYW++6d9t8fHlxMtbWYJUMW51BAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T17:50:01.809349Z","bundle_sha256":"f41289416309ea996ff1c9538f3fb9155a022f03ee601dd7026e1c8fe5689265"}}