{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NPLBQPX6W6JWO73HTBRPGT3PGW","short_pith_number":"pith:NPLBQPX6","schema_version":"1.0","canonical_sha256":"6bd6183efeb793677f679862f34f6f35b3db046887d7872064feb0ebc72e5aa3","source":{"kind":"arxiv","id":"2506.19726","version":2},"attestation_state":"computed","paper":{"title":"Precise Bayesian Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Carlos Stein Brito","submitted_at":"2025-06-24T15:42:00Z","abstract_excerpt":"Despite its long history, Bayesian neural networks (BNNs) and variational training remain underused in practice: standard Gaussian posteriors misalign with network geometry, KL terms can be brittle in high dimensions, and implementations often add complexity without reliably improving uncertainty. We revisit the problem through the lens of normalization. Because normalization layers neutralize the influence of weight magnitude, we model uncertainty \\emph{only in weight directions} using a von Mises-Fisher posterior on the unit sphere. High-dimensional geometry then yields a single, interpretab"},"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":"2506.19726","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-24T15:42:00Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"7074230ed12c8025a1a8627f1b5b44f94eb510373e59fab24f6cbf1f4df2fd35","abstract_canon_sha256":"7cb0df39877c9101a9b87d01e569d7b4d572287265f3c2672c1d98a58518cb9c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:15.557329Z","signature_b64":"dI5CWmLkzjhZEn9ACK/vz/0GEEfpNa52jKlWSgLfV69Rk8qvZmiVFRml+4hudNYXlyrdcD41+Ba65dQcE1apAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6bd6183efeb793677f679862f34f6f35b3db046887d7872064feb0ebc72e5aa3","last_reissued_at":"2026-07-05T12:06:15.556679Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:15.556679Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Precise Bayesian Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Carlos Stein Brito","submitted_at":"2025-06-24T15:42:00Z","abstract_excerpt":"Despite its long history, Bayesian neural networks (BNNs) and variational training remain underused in practice: standard Gaussian posteriors misalign with network geometry, KL terms can be brittle in high dimensions, and implementations often add complexity without reliably improving uncertainty. We revisit the problem through the lens of normalization. Because normalization layers neutralize the influence of weight magnitude, we model uncertainty \\emph{only in weight directions} using a von Mises-Fisher posterior on the unit sphere. High-dimensional geometry then yields a single, interpretab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19726","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/2506.19726/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":"2506.19726","created_at":"2026-07-05T12:06:15.556818+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.19726v2","created_at":"2026-07-05T12:06:15.556818+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19726","created_at":"2026-07-05T12:06:15.556818+00:00"},{"alias_kind":"pith_short_12","alias_value":"NPLBQPX6W6JW","created_at":"2026-07-05T12:06:15.556818+00:00"},{"alias_kind":"pith_short_16","alias_value":"NPLBQPX6W6JWO73H","created_at":"2026-07-05T12:06:15.556818+00:00"},{"alias_kind":"pith_short_8","alias_value":"NPLBQPX6","created_at":"2026-07-05T12:06:15.556818+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/NPLBQPX6W6JWO73HTBRPGT3PGW","json":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW.json","graph_json":"https://pith.science/api/pith-number/NPLBQPX6W6JWO73HTBRPGT3PGW/graph.json","events_json":"https://pith.science/api/pith-number/NPLBQPX6W6JWO73HTBRPGT3PGW/events.json","paper":"https://pith.science/paper/NPLBQPX6"},"agent_actions":{"view_html":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW","download_json":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW.json","view_paper":"https://pith.science/paper/NPLBQPX6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.19726&json=true","fetch_graph":"https://pith.science/api/pith-number/NPLBQPX6W6JWO73HTBRPGT3PGW/graph.json","fetch_events":"https://pith.science/api/pith-number/NPLBQPX6W6JWO73HTBRPGT3PGW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW/action/storage_attestation","attest_author":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW/action/author_attestation","sign_citation":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW/action/citation_signature","submit_replication":"https://pith.science/pith/NPLBQPX6W6JWO73HTBRPGT3PGW/action/replication_record"}},"created_at":"2026-07-05T12:06:15.556818+00:00","updated_at":"2026-07-05T12:06:15.556818+00:00"}