{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:7TPPYHMXIYUYAFFQTN2G7BRIEK","short_pith_number":"pith:7TPPYHMX","schema_version":"1.0","canonical_sha256":"fcdefc1d9746298014b09b746f8628228efde89e5689bfe2d6bbacad141a7a6a","source":{"kind":"arxiv","id":"2306.07158","version":1},"attestation_state":"computed","paper":{"title":"Riemannian Laplace approximations for Bayesian neural networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Federico Bergamin, Georgios Arvanitidis, Pablo Moreno-Mu\\~noz, S{\\o}ren Hauberg","submitted_at":"2023-06-12T14:44:22Z","abstract_excerpt":"Bayesian neural networks often approximate the weight-posterior with a Gaussian distribution. However, practical posteriors are often, even locally, highly non-Gaussian, and empirical performance deteriorates. We propose a simple parametric approximate posterior that adapts to the shape of the true posterior through a Riemannian metric that is determined by the log-posterior gradient. We develop a Riemannian Laplace approximation where samples naturally fall into weight-regions with low negative log-posterior. We show that these samples can be drawn by solving a system of ordinary differential"},"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":"2306.07158","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-06-12T14:44:22Z","cross_cats_sorted":["cs.LG","stat.ME"],"title_canon_sha256":"0d96fa366564f42e93582445eed6f6f3960237ffbdc815c8f71aff30096e8fa3","abstract_canon_sha256":"d24c60442ee79d419f66a93e817e22b3fe402c40d86e548bf815a2b8195cfc8d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:19:48.453717Z","signature_b64":"JkDAP103Gm9y/56pEdfmFouytNVDMW39EZIBm7eUEEFftG7kwQU2m8HAQfwGXh+bsTysHlnY4353dhZrmCcNCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcdefc1d9746298014b09b746f8628228efde89e5689bfe2d6bbacad141a7a6a","last_reissued_at":"2026-07-05T06:19:48.453182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:19:48.453182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Riemannian Laplace approximations for Bayesian neural networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Federico Bergamin, Georgios Arvanitidis, Pablo Moreno-Mu\\~noz, S{\\o}ren Hauberg","submitted_at":"2023-06-12T14:44:22Z","abstract_excerpt":"Bayesian neural networks often approximate the weight-posterior with a Gaussian distribution. However, practical posteriors are often, even locally, highly non-Gaussian, and empirical performance deteriorates. We propose a simple parametric approximate posterior that adapts to the shape of the true posterior through a Riemannian metric that is determined by the log-posterior gradient. We develop a Riemannian Laplace approximation where samples naturally fall into weight-regions with low negative log-posterior. We show that these samples can be drawn by solving a system of ordinary differential"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07158","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/2306.07158/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":"2306.07158","created_at":"2026-07-05T06:19:48.453238+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07158v1","created_at":"2026-07-05T06:19:48.453238+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07158","created_at":"2026-07-05T06:19:48.453238+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TPPYHMXIYUY","created_at":"2026-07-05T06:19:48.453238+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TPPYHMXIYUYAFFQ","created_at":"2026-07-05T06:19:48.453238+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TPPYHMX","created_at":"2026-07-05T06:19:48.453238+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/7TPPYHMXIYUYAFFQTN2G7BRIEK","json":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK.json","graph_json":"https://pith.science/api/pith-number/7TPPYHMXIYUYAFFQTN2G7BRIEK/graph.json","events_json":"https://pith.science/api/pith-number/7TPPYHMXIYUYAFFQTN2G7BRIEK/events.json","paper":"https://pith.science/paper/7TPPYHMX"},"agent_actions":{"view_html":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK","download_json":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK.json","view_paper":"https://pith.science/paper/7TPPYHMX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07158&json=true","fetch_graph":"https://pith.science/api/pith-number/7TPPYHMXIYUYAFFQTN2G7BRIEK/graph.json","fetch_events":"https://pith.science/api/pith-number/7TPPYHMXIYUYAFFQTN2G7BRIEK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK/action/storage_attestation","attest_author":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK/action/author_attestation","sign_citation":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK/action/citation_signature","submit_replication":"https://pith.science/pith/7TPPYHMXIYUYAFFQTN2G7BRIEK/action/replication_record"}},"created_at":"2026-07-05T06:19:48.453238+00:00","updated_at":"2026-07-05T06:19:48.453238+00:00"}