{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:HWU7D357VGJ4HAPZUJARAGY2ZI","short_pith_number":"pith:HWU7D357","schema_version":"1.0","canonical_sha256":"3da9f1efbfa993c381f9a241101b1aca199f3f72c2127a3b4ad061cec5b0f487","source":{"kind":"arxiv","id":"2209.15618","version":2},"attestation_state":"computed","paper":{"title":"Beyond Bayes-optimality: meta-learning what you know you don't know","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anian Ruoss, Chris Cundy, Christopher Summerfield, Elliot Catt, Gr\\'egoire Del\\'etang, Jane Wang, Joel Veness, Jordi Grau-Moya, Kevin Li, Marcus Hutter, Markus Kunesch, Pedro Ortega, Shane Legg, Tim Genewein","submitted_at":"2022-09-30T17:44:01Z","abstract_excerpt":"Meta-training agents with memory has been shown to culminate in Bayes-optimal agents, which casts Bayes-optimality as the implicit solution to a numerical optimization problem rather than an explicit modeling assumption. Bayes-optimal agents are risk-neutral, since they solely attune to the expected return, and ambiguity-neutral, since they act in new situations as if the uncertainty were known. This is in contrast to risk-sensitive agents, which additionally exploit the higher-order moments of the return, and ambiguity-sensitive agents, which act differently when recognizing situations in whi"},"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":"2209.15618","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2022-09-30T17:44:01Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7b775c4cd36e5baa024f41d60bc41a6bcc2d79ed3918ebce18b01aafd827e7b2","abstract_canon_sha256":"1d5e6901edd909adbbd1913390600434765eabb723070eaea6b034fb7c040594"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:05:50.902096Z","signature_b64":"3AEIQKMLWpIAutJQyVad26VAM9q20aGdKwp7lxPLKU55wGTyCpSzH4C+HkzA7CWqLpRJrGlOb+0GZ8MD/yYfBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3da9f1efbfa993c381f9a241101b1aca199f3f72c2127a3b4ad061cec5b0f487","last_reissued_at":"2026-07-05T05:05:50.901651Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:05:50.901651Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Bayes-optimality: meta-learning what you know you don't know","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anian Ruoss, Chris Cundy, Christopher Summerfield, Elliot Catt, Gr\\'egoire Del\\'etang, Jane Wang, Joel Veness, Jordi Grau-Moya, Kevin Li, Marcus Hutter, Markus Kunesch, Pedro Ortega, Shane Legg, Tim Genewein","submitted_at":"2022-09-30T17:44:01Z","abstract_excerpt":"Meta-training agents with memory has been shown to culminate in Bayes-optimal agents, which casts Bayes-optimality as the implicit solution to a numerical optimization problem rather than an explicit modeling assumption. Bayes-optimal agents are risk-neutral, since they solely attune to the expected return, and ambiguity-neutral, since they act in new situations as if the uncertainty were known. This is in contrast to risk-sensitive agents, which additionally exploit the higher-order moments of the return, and ambiguity-sensitive agents, which act differently when recognizing situations in whi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.15618","kind":"arxiv","version":2},"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/2209.15618/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":"2209.15618","created_at":"2026-07-05T05:05:50.901712+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.15618v2","created_at":"2026-07-05T05:05:50.901712+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.15618","created_at":"2026-07-05T05:05:50.901712+00:00"},{"alias_kind":"pith_short_12","alias_value":"HWU7D357VGJ4","created_at":"2026-07-05T05:05:50.901712+00:00"},{"alias_kind":"pith_short_16","alias_value":"HWU7D357VGJ4HAPZ","created_at":"2026-07-05T05:05:50.901712+00:00"},{"alias_kind":"pith_short_8","alias_value":"HWU7D357","created_at":"2026-07-05T05:05:50.901712+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.16288","citing_title":"Next-Token Prediction Should be Ambiguity-Sensitive: A Meta-Learning Perspective","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI","json":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI.json","graph_json":"https://pith.science/api/pith-number/HWU7D357VGJ4HAPZUJARAGY2ZI/graph.json","events_json":"https://pith.science/api/pith-number/HWU7D357VGJ4HAPZUJARAGY2ZI/events.json","paper":"https://pith.science/paper/HWU7D357"},"agent_actions":{"view_html":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI","download_json":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI.json","view_paper":"https://pith.science/paper/HWU7D357","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.15618&json=true","fetch_graph":"https://pith.science/api/pith-number/HWU7D357VGJ4HAPZUJARAGY2ZI/graph.json","fetch_events":"https://pith.science/api/pith-number/HWU7D357VGJ4HAPZUJARAGY2ZI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI/action/storage_attestation","attest_author":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI/action/author_attestation","sign_citation":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI/action/citation_signature","submit_replication":"https://pith.science/pith/HWU7D357VGJ4HAPZUJARAGY2ZI/action/replication_record"}},"created_at":"2026-07-05T05:05:50.901712+00:00","updated_at":"2026-07-05T05:05:50.901712+00:00"}