{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:PSDKYXBUBJZ537SK35J4TOYPKL","short_pith_number":"pith:PSDKYXBU","canonical_record":{"source":{"id":"2201.09802","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-24T17:02:22Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"caf331b9e7c0ee43e8c73ab6eccf41dd28025a4359682c6c2a386d4f109071d7","abstract_canon_sha256":"18529beb45d5f7e23ae5c06afa616d9c7748190d05961cb12107125a7dd74925"},"schema_version":"1.0"},"canonical_sha256":"7c86ac5c340a73ddfe4adf53c9bb0f52ef64b340f945e8603dadd1d09c7c2f27","source":{"kind":"arxiv","id":"2201.09802","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.09802","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"arxiv_version","alias_value":"2201.09802v4","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.09802","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"pith_short_12","alias_value":"PSDKYXBUBJZ5","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"pith_short_16","alias_value":"PSDKYXBUBJZ537SK","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"pith_short_8","alias_value":"PSDKYXBU","created_at":"2026-07-05T03:54:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:PSDKYXBUBJZ537SK35J4TOYPKL","target":"record","payload":{"canonical_record":{"source":{"id":"2201.09802","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-24T17:02:22Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"caf331b9e7c0ee43e8c73ab6eccf41dd28025a4359682c6c2a386d4f109071d7","abstract_canon_sha256":"18529beb45d5f7e23ae5c06afa616d9c7748190d05961cb12107125a7dd74925"},"schema_version":"1.0"},"canonical_sha256":"7c86ac5c340a73ddfe4adf53c9bb0f52ef64b340f945e8603dadd1d09c7c2f27","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:54:27.008154Z","signature_b64":"MQXb/HeXAB60ouvmpnsXdZo7iYL1r0Dsq0AYQScnXJcBz4NUoMi601k9FM6UD37/mN1pps5JHYnnIQO0Zy36Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c86ac5c340a73ddfe4adf53c9bb0f52ef64b340f945e8603dadd1d09c7c2f27","last_reissued_at":"2026-07-05T03:54:27.007774Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:54:27.007774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2201.09802","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:54:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N/bc/gmITAOuwzyryTLCkT/VcKY6jwfZtcgHA08yFqdWNFXXhLt/UDcVFfyDvgoJM2ySAhgLzQW7X9xBSBklAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T05:29:35.857365Z"},"content_sha256":"ed1263f0192ba605d933f6e1a2c2792cf6d4373702cdb9e7e5c7c017ef97e944","schema_version":"1.0","event_id":"sha256:ed1263f0192ba605d933f6e1a2c2792cf6d4373702cdb9e7e5c7c017ef97e944"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:PSDKYXBUBJZ537SK35J4TOYPKL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Constrained Policy Optimization via Bayesian World Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Andreas Krause, Ilnura Usmanova, Sebastian Curi, Yarden As","submitted_at":"2022-01-24T17:02:22Z","abstract_excerpt":"Improving sample-efficiency and safety are crucial challenges when deploying reinforcement learning in high-stakes real world applications. We propose LAMBDA, a novel model-based approach for policy optimization in safety critical tasks modeled via constrained Markov decision processes. Our approach utilizes Bayesian world models, and harnesses the resulting uncertainty to maximize optimistic upper bounds on the task objective, as well as pessimistic upper bounds on the safety constraints. We demonstrate LAMBDA's state of the art performance on the Safety-Gym benchmark suite in terms of sample"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.09802","kind":"arxiv","version":4},"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/2201.09802/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":"2026-07-05T03:54:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"M1ckSJAojrXLUphuet3+73jm02qs/0F6KKE7NYLGNFbKKHxZ287rm9jNct5IYJwj4+c61gPBKs+h3cDAr64TAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T05:29:35.857872Z"},"content_sha256":"31e6d4ef70e8543813b2b3ccaf74c31d07e1f3419761a0ae853fa127b216295b","schema_version":"1.0","event_id":"sha256:31e6d4ef70e8543813b2b3ccaf74c31d07e1f3419761a0ae853fa127b216295b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PSDKYXBUBJZ537SK35J4TOYPKL/bundle.json","state_url":"https://pith.science/pith/PSDKYXBUBJZ537SK35J4TOYPKL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PSDKYXBUBJZ537SK35J4TOYPKL/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":"2026-08-09T05:29:35Z","links":{"resolver":"https://pith.science/pith/PSDKYXBUBJZ537SK35J4TOYPKL","bundle":"https://pith.science/pith/PSDKYXBUBJZ537SK35J4TOYPKL/bundle.json","state":"https://pith.science/pith/PSDKYXBUBJZ537SK35J4TOYPKL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PSDKYXBUBJZ537SK35J4TOYPKL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:PSDKYXBUBJZ537SK35J4TOYPKL","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":"18529beb45d5f7e23ae5c06afa616d9c7748190d05961cb12107125a7dd74925","cross_cats_sorted":["cs.AI","cs.RO"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-24T17:02:22Z","title_canon_sha256":"caf331b9e7c0ee43e8c73ab6eccf41dd28025a4359682c6c2a386d4f109071d7"},"schema_version":"1.0","source":{"id":"2201.09802","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.09802","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"arxiv_version","alias_value":"2201.09802v4","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.09802","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"pith_short_12","alias_value":"PSDKYXBUBJZ5","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"pith_short_16","alias_value":"PSDKYXBUBJZ537SK","created_at":"2026-07-05T03:54:27Z"},{"alias_kind":"pith_short_8","alias_value":"PSDKYXBU","created_at":"2026-07-05T03:54:27Z"}],"graph_snapshots":[{"event_id":"sha256:31e6d4ef70e8543813b2b3ccaf74c31d07e1f3419761a0ae853fa127b216295b","target":"graph","created_at":"2026-07-05T03:54:27Z","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/2201.09802/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Improving sample-efficiency and safety are crucial challenges when deploying reinforcement learning in high-stakes real world applications. We propose LAMBDA, a novel model-based approach for policy optimization in safety critical tasks modeled via constrained Markov decision processes. Our approach utilizes Bayesian world models, and harnesses the resulting uncertainty to maximize optimistic upper bounds on the task objective, as well as pessimistic upper bounds on the safety constraints. We demonstrate LAMBDA's state of the art performance on the Safety-Gym benchmark suite in terms of sample","authors_text":"Andreas Krause, Ilnura Usmanova, Sebastian Curi, Yarden As","cross_cats":["cs.AI","cs.RO"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-24T17:02:22Z","title":"Constrained Policy Optimization via Bayesian World Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.09802","kind":"arxiv","version":4},"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:ed1263f0192ba605d933f6e1a2c2792cf6d4373702cdb9e7e5c7c017ef97e944","target":"record","created_at":"2026-07-05T03:54:27Z","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":"18529beb45d5f7e23ae5c06afa616d9c7748190d05961cb12107125a7dd74925","cross_cats_sorted":["cs.AI","cs.RO"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-24T17:02:22Z","title_canon_sha256":"caf331b9e7c0ee43e8c73ab6eccf41dd28025a4359682c6c2a386d4f109071d7"},"schema_version":"1.0","source":{"id":"2201.09802","kind":"arxiv","version":4}},"canonical_sha256":"7c86ac5c340a73ddfe4adf53c9bb0f52ef64b340f945e8603dadd1d09c7c2f27","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7c86ac5c340a73ddfe4adf53c9bb0f52ef64b340f945e8603dadd1d09c7c2f27","first_computed_at":"2026-07-05T03:54:27.007774Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:54:27.007774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MQXb/HeXAB60ouvmpnsXdZo7iYL1r0Dsq0AYQScnXJcBz4NUoMi601k9FM6UD37/mN1pps5JHYnnIQO0Zy36Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T03:54:27.008154Z","signed_message":"canonical_sha256_bytes"},"source_id":"2201.09802","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ed1263f0192ba605d933f6e1a2c2792cf6d4373702cdb9e7e5c7c017ef97e944","sha256:31e6d4ef70e8543813b2b3ccaf74c31d07e1f3419761a0ae853fa127b216295b"],"state_sha256":"9d3be6043e7735dbe87760308d62b27f60ca2236d4d5b4e48ed8fd02aa647bce"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qiyNGjus1xPhzZXWgJTWZFSPHrXq1FV4iJH81rjPRzLTn2SjxnTFu4PX/U8h8+fl9rqO/TgNr1a7eitgYiNcBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T05:29:35.863193Z","bundle_sha256":"ebbc5458080ff77435e1250256fd37a4002cca677dc6b0c3a3507d62452402ea"}}