{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:YGAJWQBKVEQGCUJXMEJI5ZSFGG","short_pith_number":"pith:YGAJWQBK","schema_version":"1.0","canonical_sha256":"c1809b402aa92061513761128ee64531a2a412d5f29401d4384bf1bf5c6cd788","source":{"kind":"arxiv","id":"2012.13220","version":1},"attestation_state":"computed","paper":{"title":"On Batch Normalisation for Approximate Bayesian Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jishnu Mukhoti, Philip H.S. Torr, Puneet K. Dokania, Yarin Gal","submitted_at":"2020-12-24T12:40:11Z","abstract_excerpt":"We study batch normalisation in the context of variational inference methods in Bayesian neural networks, such as mean-field or MC Dropout. We show that batch-normalisation does not affect the optimum of the evidence lower bound (ELBO). Furthermore, we study the Monte Carlo Batch Normalisation (MCBN) algorithm, proposed as an approximate inference technique parallel to MC Dropout, and show that for larger batch sizes, MCBN fails to capture epistemic uncertainty. Finally, we provide insights into what is required to fix this failure, namely having to view the mini-batch size as a variational pa"},"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":"2012.13220","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-12-24T12:40:11Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"af39f3510d63372c6a2f6efb14bb3c29e569dd7abaab1415420a098d6e865892","abstract_canon_sha256":"670c097d14cdb0b5baedee9b76ebcf90960b0cbf3c38a987e0bc13fb2a7d84eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:01:56.400930Z","signature_b64":"K5K2E7Bbo5o1W+i0lZ3Y60D9m4CICP/Y4fcypy5o5bGVJ997hku5N3uWlj23M9NfhPCYwVyzcRp/f8eQ9I3rCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1809b402aa92061513761128ee64531a2a412d5f29401d4384bf1bf5c6cd788","last_reissued_at":"2026-07-05T02:01:56.400519Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:01:56.400519Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Batch Normalisation for Approximate Bayesian Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jishnu Mukhoti, Philip H.S. Torr, Puneet K. Dokania, Yarin Gal","submitted_at":"2020-12-24T12:40:11Z","abstract_excerpt":"We study batch normalisation in the context of variational inference methods in Bayesian neural networks, such as mean-field or MC Dropout. We show that batch-normalisation does not affect the optimum of the evidence lower bound (ELBO). Furthermore, we study the Monte Carlo Batch Normalisation (MCBN) algorithm, proposed as an approximate inference technique parallel to MC Dropout, and show that for larger batch sizes, MCBN fails to capture epistemic uncertainty. Finally, we provide insights into what is required to fix this failure, namely having to view the mini-batch size as a variational pa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.13220","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/2012.13220/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":"2012.13220","created_at":"2026-07-05T02:01:56.400575+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.13220v1","created_at":"2026-07-05T02:01:56.400575+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.13220","created_at":"2026-07-05T02:01:56.400575+00:00"},{"alias_kind":"pith_short_12","alias_value":"YGAJWQBKVEQG","created_at":"2026-07-05T02:01:56.400575+00:00"},{"alias_kind":"pith_short_16","alias_value":"YGAJWQBKVEQGCUJX","created_at":"2026-07-05T02:01:56.400575+00:00"},{"alias_kind":"pith_short_8","alias_value":"YGAJWQBK","created_at":"2026-07-05T02:01:56.400575+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/YGAJWQBKVEQGCUJXMEJI5ZSFGG","json":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG.json","graph_json":"https://pith.science/api/pith-number/YGAJWQBKVEQGCUJXMEJI5ZSFGG/graph.json","events_json":"https://pith.science/api/pith-number/YGAJWQBKVEQGCUJXMEJI5ZSFGG/events.json","paper":"https://pith.science/paper/YGAJWQBK"},"agent_actions":{"view_html":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG","download_json":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG.json","view_paper":"https://pith.science/paper/YGAJWQBK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.13220&json=true","fetch_graph":"https://pith.science/api/pith-number/YGAJWQBKVEQGCUJXMEJI5ZSFGG/graph.json","fetch_events":"https://pith.science/api/pith-number/YGAJWQBKVEQGCUJXMEJI5ZSFGG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG/action/storage_attestation","attest_author":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG/action/author_attestation","sign_citation":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG/action/citation_signature","submit_replication":"https://pith.science/pith/YGAJWQBKVEQGCUJXMEJI5ZSFGG/action/replication_record"}},"created_at":"2026-07-05T02:01:56.400575+00:00","updated_at":"2026-07-05T02:01:56.400575+00:00"}