{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:VDPDDWQZRLHME7WLCJPSPPOWIH","short_pith_number":"pith:VDPDDWQZ","schema_version":"1.0","canonical_sha256":"a8de31da198acec27ecb125f27bdd641ea75b688b5f7c43ee994bff2a320c75a","source":{"kind":"arxiv","id":"2302.09656","version":5},"attestation_state":"computed","paper":{"title":"Credal Bayesian Deep Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Insup Lee, Kuk Jin Jang, Michele Caprio, Oleg Sokolsky, Radoslav Ivanov, Souradeep Dutta, Vivian Lin","submitted_at":"2023-02-19T19:03:26Z","abstract_excerpt":"Uncertainty quantification and robustness to distribution shifts are important goals in machine learning and artificial intelligence. Although Bayesian Neural Networks (BNNs) allow for uncertainty in the predictions to be assessed, different sources of predictive uncertainty cannot be distinguished properly. We present Credal Bayesian Deep Learning (CBDL). Heuristically, CBDL allows to train an (uncountably) infinite ensemble of BNNs, using only finitely many elements. This is possible thanks to prior and likelihood finitely generated credal sets (FGCSs), a concept from the imprecise probabili"},"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":"2302.09656","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-19T19:03:26Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"3bd86133bebda118dca9488daa2157d58ee5c05586ce5595cb12418d0619bcae","abstract_canon_sha256":"6bf5a456eda38baaf4b1e97f4cf6115c6bfa2e9287def9856f16547d34361a71"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:23:43.035512Z","signature_b64":"IG07iA6OlDNMECDzMEg7Chd6kS3sxd2STnhjTudji1QgnKBiPS9bJ/LkPKs5r1R1p6WZZNzTW/u9TOqxHX54AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a8de31da198acec27ecb125f27bdd641ea75b688b5f7c43ee994bff2a320c75a","last_reissued_at":"2026-07-05T09:23:43.035066Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:23:43.035066Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Credal Bayesian Deep Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Insup Lee, Kuk Jin Jang, Michele Caprio, Oleg Sokolsky, Radoslav Ivanov, Souradeep Dutta, Vivian Lin","submitted_at":"2023-02-19T19:03:26Z","abstract_excerpt":"Uncertainty quantification and robustness to distribution shifts are important goals in machine learning and artificial intelligence. Although Bayesian Neural Networks (BNNs) allow for uncertainty in the predictions to be assessed, different sources of predictive uncertainty cannot be distinguished properly. We present Credal Bayesian Deep Learning (CBDL). Heuristically, CBDL allows to train an (uncountably) infinite ensemble of BNNs, using only finitely many elements. This is possible thanks to prior and likelihood finitely generated credal sets (FGCSs), a concept from the imprecise probabili"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.09656","kind":"arxiv","version":5},"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/2302.09656/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":"2302.09656","created_at":"2026-07-05T09:23:43.035123+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.09656v5","created_at":"2026-07-05T09:23:43.035123+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.09656","created_at":"2026-07-05T09:23:43.035123+00:00"},{"alias_kind":"pith_short_12","alias_value":"VDPDDWQZRLHM","created_at":"2026-07-05T09:23:43.035123+00:00"},{"alias_kind":"pith_short_16","alias_value":"VDPDDWQZRLHME7WL","created_at":"2026-07-05T09:23:43.035123+00:00"},{"alias_kind":"pith_short_8","alias_value":"VDPDDWQZ","created_at":"2026-07-05T09:23:43.035123+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25882","citing_title":"Conformalised imprecise inference for robust extrapolation under limited data","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH","json":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH.json","graph_json":"https://pith.science/api/pith-number/VDPDDWQZRLHME7WLCJPSPPOWIH/graph.json","events_json":"https://pith.science/api/pith-number/VDPDDWQZRLHME7WLCJPSPPOWIH/events.json","paper":"https://pith.science/paper/VDPDDWQZ"},"agent_actions":{"view_html":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH","download_json":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH.json","view_paper":"https://pith.science/paper/VDPDDWQZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.09656&json=true","fetch_graph":"https://pith.science/api/pith-number/VDPDDWQZRLHME7WLCJPSPPOWIH/graph.json","fetch_events":"https://pith.science/api/pith-number/VDPDDWQZRLHME7WLCJPSPPOWIH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH/action/storage_attestation","attest_author":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH/action/author_attestation","sign_citation":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH/action/citation_signature","submit_replication":"https://pith.science/pith/VDPDDWQZRLHME7WLCJPSPPOWIH/action/replication_record"}},"created_at":"2026-07-05T09:23:43.035123+00:00","updated_at":"2026-07-05T09:23:43.035123+00:00"}