{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:JBP2HCEXSXRQYMAH6FM74PWF3U","short_pith_number":"pith:JBP2HCEX","canonical_record":{"source":{"id":"2104.09469","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-19T17:33:07Z","cross_cats_sorted":["cs.AI","cs.HC"],"title_canon_sha256":"616a32c5adcfb01e2b5cb64fb8e3bf73ccbe4034e27e6840212fd82d7149832c","abstract_canon_sha256":"979c2453543142ad562c2d9413fd1a585c290d69f3245a864e496b16afff0916"},"schema_version":"1.0"},"canonical_sha256":"485fa3889795e30c3007f159fe3ec5dd0ff0d332a4bb534ec9b2d373a8497aea","source":{"kind":"arxiv","id":"2104.09469","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.09469","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"arxiv_version","alias_value":"2104.09469v1","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.09469","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"pith_short_12","alias_value":"JBP2HCEXSXRQ","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"pith_short_16","alias_value":"JBP2HCEXSXRQYMAH","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"pith_short_8","alias_value":"JBP2HCEX","created_at":"2026-07-05T02:33:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:JBP2HCEXSXRQYMAH6FM74PWF3U","target":"record","payload":{"canonical_record":{"source":{"id":"2104.09469","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-19T17:33:07Z","cross_cats_sorted":["cs.AI","cs.HC"],"title_canon_sha256":"616a32c5adcfb01e2b5cb64fb8e3bf73ccbe4034e27e6840212fd82d7149832c","abstract_canon_sha256":"979c2453543142ad562c2d9413fd1a585c290d69f3245a864e496b16afff0916"},"schema_version":"1.0"},"canonical_sha256":"485fa3889795e30c3007f159fe3ec5dd0ff0d332a4bb534ec9b2d373a8497aea","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:33:13.592462Z","signature_b64":"DBs9tk803iA0Fk+DbCpq7aAVIWowkTpmoMBFOHq3GIQuiu7TIKyDZYlBAC6k79LDNVaWd0afmH0mVpMiGJBCBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"485fa3889795e30c3007f159fe3ec5dd0ff0d332a4bb534ec9b2d373a8497aea","last_reissued_at":"2026-07-05T02:33:13.591986Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:33:13.591986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2104.09469","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":"2026-07-05T02:33:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cO+ILowbTYzoEHKFex7gcga1FrMsCkZXy/ZLTm0/lHAVQoQerbsDXFLpN2aWNcGifBnRVjGnHRQa29XvR80vCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T09:59:53.086852Z"},"content_sha256":"664cf45269be8adcd99369e560d3f7420da46de096021c3dc9b9983809843d14","schema_version":"1.0","event_id":"sha256:664cf45269be8adcd99369e560d3f7420da46de096021c3dc9b9983809843d14"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:JBP2HCEXSXRQYMAH6FM74PWF3U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training Value-Aligned Reinforcement Learning Agents Using a Normative Prior","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC"],"primary_cat":"cs.LG","authors_text":"Brent Harrison, Mark Riedl, Md Sultan Al Nahian, Spencer Frazier","submitted_at":"2021-04-19T17:33:07Z","abstract_excerpt":"As more machine learning agents interact with humans, it is increasingly a prospect that an agent trained to perform a task optimally, using only a measure of task performance as feedback, can violate societal norms for acceptable behavior or cause harm. Value alignment is a property of intelligent agents wherein they solely pursue non-harmful behaviors or human-beneficial goals. We introduce an approach to value-aligned reinforcement learning, in which we train an agent with two reward signals: a standard task performance reward, plus a normative behavior reward. The normative behavior reward"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.09469","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/2104.09469/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":"2026-07-05T02:33:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5dVkzq9ur1JBBUG4vOF3mTYGs5GGEO8Xdglicl55kcXy8KHH7IvkjOe3CcdyvumBqfg3SL2qPY5Oi4h/jgDDCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T09:59:53.087374Z"},"content_sha256":"385923652bca2692b4b73febe8f90a35de53478d95d90f760345c582edbafd01","schema_version":"1.0","event_id":"sha256:385923652bca2692b4b73febe8f90a35de53478d95d90f760345c582edbafd01"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JBP2HCEXSXRQYMAH6FM74PWF3U/bundle.json","state_url":"https://pith.science/pith/JBP2HCEXSXRQYMAH6FM74PWF3U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JBP2HCEXSXRQYMAH6FM74PWF3U/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":"2026-08-10T09:59:53Z","links":{"resolver":"https://pith.science/pith/JBP2HCEXSXRQYMAH6FM74PWF3U","bundle":"https://pith.science/pith/JBP2HCEXSXRQYMAH6FM74PWF3U/bundle.json","state":"https://pith.science/pith/JBP2HCEXSXRQYMAH6FM74PWF3U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JBP2HCEXSXRQYMAH6FM74PWF3U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:JBP2HCEXSXRQYMAH6FM74PWF3U","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":"979c2453543142ad562c2d9413fd1a585c290d69f3245a864e496b16afff0916","cross_cats_sorted":["cs.AI","cs.HC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-19T17:33:07Z","title_canon_sha256":"616a32c5adcfb01e2b5cb64fb8e3bf73ccbe4034e27e6840212fd82d7149832c"},"schema_version":"1.0","source":{"id":"2104.09469","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.09469","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"arxiv_version","alias_value":"2104.09469v1","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.09469","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"pith_short_12","alias_value":"JBP2HCEXSXRQ","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"pith_short_16","alias_value":"JBP2HCEXSXRQYMAH","created_at":"2026-07-05T02:33:13Z"},{"alias_kind":"pith_short_8","alias_value":"JBP2HCEX","created_at":"2026-07-05T02:33:13Z"}],"graph_snapshots":[{"event_id":"sha256:385923652bca2692b4b73febe8f90a35de53478d95d90f760345c582edbafd01","target":"graph","created_at":"2026-07-05T02:33:13Z","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/2104.09469/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As more machine learning agents interact with humans, it is increasingly a prospect that an agent trained to perform a task optimally, using only a measure of task performance as feedback, can violate societal norms for acceptable behavior or cause harm. Value alignment is a property of intelligent agents wherein they solely pursue non-harmful behaviors or human-beneficial goals. We introduce an approach to value-aligned reinforcement learning, in which we train an agent with two reward signals: a standard task performance reward, plus a normative behavior reward. The normative behavior reward","authors_text":"Brent Harrison, Mark Riedl, Md Sultan Al Nahian, Spencer Frazier","cross_cats":["cs.AI","cs.HC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-19T17:33:07Z","title":"Training Value-Aligned Reinforcement Learning Agents Using a Normative Prior"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.09469","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":"sha256:664cf45269be8adcd99369e560d3f7420da46de096021c3dc9b9983809843d14","target":"record","created_at":"2026-07-05T02:33:13Z","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":"979c2453543142ad562c2d9413fd1a585c290d69f3245a864e496b16afff0916","cross_cats_sorted":["cs.AI","cs.HC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-19T17:33:07Z","title_canon_sha256":"616a32c5adcfb01e2b5cb64fb8e3bf73ccbe4034e27e6840212fd82d7149832c"},"schema_version":"1.0","source":{"id":"2104.09469","kind":"arxiv","version":1}},"canonical_sha256":"485fa3889795e30c3007f159fe3ec5dd0ff0d332a4bb534ec9b2d373a8497aea","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"485fa3889795e30c3007f159fe3ec5dd0ff0d332a4bb534ec9b2d373a8497aea","first_computed_at":"2026-07-05T02:33:13.591986Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:33:13.591986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DBs9tk803iA0Fk+DbCpq7aAVIWowkTpmoMBFOHq3GIQuiu7TIKyDZYlBAC6k79LDNVaWd0afmH0mVpMiGJBCBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:33:13.592462Z","signed_message":"canonical_sha256_bytes"},"source_id":"2104.09469","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:664cf45269be8adcd99369e560d3f7420da46de096021c3dc9b9983809843d14","sha256:385923652bca2692b4b73febe8f90a35de53478d95d90f760345c582edbafd01"],"state_sha256":"0febd7d1c68904533649608daf309f0291ae79f1692e1e80ed4ae348f6c2f24c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"O2qP7SOC03MznSE2ruPJXM0aDzX0CmrvWNRdl09lE2ekhpd6qo7BIfV27Z4qWl3ssfBqaX+kn1oQUZmKxzs2Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T09:59:53.091632Z","bundle_sha256":"bf15351789bdd111569e9239e0fa8c6d0d2e812d5e99a5c2de80c1a7cb46d4c7"}}