{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:OX4APUAJ5Y4XCXBCIGWMPU5YAB","short_pith_number":"pith:OX4APUAJ","schema_version":"1.0","canonical_sha256":"75f807d009ee39715c2241acc7d3b800755366bd5df41e86f8d9832d0756ffbf","source":{"kind":"arxiv","id":"2106.07052","version":1},"attestation_state":"computed","paper":{"title":"Wide Mean-Field Variational Bayesian Neural Networks Ignore the Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Beau Coker, Finale Doshi-Velez, Weiwei Pan","submitted_at":"2021-06-13T17:36:38Z","abstract_excerpt":"Variational inference enables approximate posterior inference of the highly over-parameterized neural networks that are popular in modern machine learning. Unfortunately, such posteriors are known to exhibit various pathological behaviors. We prove that as the number of hidden units in a single-layer Bayesian neural network tends to infinity, the function-space posterior mean under mean-field variational inference actually converges to zero, completely ignoring the data. This is in contrast to the true posterior, which converges to a Gaussian process. Our work provides insight into the over-re"},"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":"2106.07052","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-13T17:36:38Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"56371d6f32a3e2655ce699a5d87d6a0b9ac865e846a7c2ed78810e5a7dc919da","abstract_canon_sha256":"ecfb0ea4a2a214c33df121e02a42bc6109bfebf400002e5428132518e204bcf7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:49:02.363828Z","signature_b64":"xJPv+t3fR1I1Vu/M97yovX3eAsTXsXG+/Hfia62BkaqvNUdh8jQ84YyXihLG917bXMCBrc68ewJNYj68HqyoDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"75f807d009ee39715c2241acc7d3b800755366bd5df41e86f8d9832d0756ffbf","last_reissued_at":"2026-07-05T02:49:02.363422Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:49:02.363422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wide Mean-Field Variational Bayesian Neural Networks Ignore the Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Beau Coker, Finale Doshi-Velez, Weiwei Pan","submitted_at":"2021-06-13T17:36:38Z","abstract_excerpt":"Variational inference enables approximate posterior inference of the highly over-parameterized neural networks that are popular in modern machine learning. Unfortunately, such posteriors are known to exhibit various pathological behaviors. We prove that as the number of hidden units in a single-layer Bayesian neural network tends to infinity, the function-space posterior mean under mean-field variational inference actually converges to zero, completely ignoring the data. This is in contrast to the true posterior, which converges to a Gaussian process. Our work provides insight into the over-re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.07052","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/2106.07052/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":"2106.07052","created_at":"2026-07-05T02:49:02.363485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.07052v1","created_at":"2026-07-05T02:49:02.363485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.07052","created_at":"2026-07-05T02:49:02.363485+00:00"},{"alias_kind":"pith_short_12","alias_value":"OX4APUAJ5Y4X","created_at":"2026-07-05T02:49:02.363485+00:00"},{"alias_kind":"pith_short_16","alias_value":"OX4APUAJ5Y4XCXBC","created_at":"2026-07-05T02:49:02.363485+00:00"},{"alias_kind":"pith_short_8","alias_value":"OX4APUAJ","created_at":"2026-07-05T02:49:02.363485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB","json":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB.json","graph_json":"https://pith.science/api/pith-number/OX4APUAJ5Y4XCXBCIGWMPU5YAB/graph.json","events_json":"https://pith.science/api/pith-number/OX4APUAJ5Y4XCXBCIGWMPU5YAB/events.json","paper":"https://pith.science/paper/OX4APUAJ"},"agent_actions":{"view_html":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB","download_json":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB.json","view_paper":"https://pith.science/paper/OX4APUAJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.07052&json=true","fetch_graph":"https://pith.science/api/pith-number/OX4APUAJ5Y4XCXBCIGWMPU5YAB/graph.json","fetch_events":"https://pith.science/api/pith-number/OX4APUAJ5Y4XCXBCIGWMPU5YAB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB/action/storage_attestation","attest_author":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB/action/author_attestation","sign_citation":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB/action/citation_signature","submit_replication":"https://pith.science/pith/OX4APUAJ5Y4XCXBCIGWMPU5YAB/action/replication_record"}},"created_at":"2026-07-05T02:49:02.363485+00:00","updated_at":"2026-07-05T02:49:02.363485+00:00"}