{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JUPRYIUFL34FXVXPBTRWGYUULG","short_pith_number":"pith:JUPRYIUF","schema_version":"1.0","canonical_sha256":"4d1f1c22855ef85bd6ef0ce363629459ac0ca5265efcd373947e0830283f3b89","source":{"kind":"arxiv","id":"2002.08791","version":4},"attestation_state":"computed","paper":{"title":"Bayesian Deep Learning and a Probabilistic Perspective of Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Gordon Wilson, Pavel Izmailov","submitted_at":"2020-02-20T15:13:27Z","abstract_excerpt":"The key distinguishing property of a Bayesian approach is marginalization, rather than using a single setting of weights. Bayesian marginalization can particularly improve the accuracy and calibration of modern deep neural networks, which are typically underspecified by the data, and can represent many compelling but different solutions. We show that deep ensembles provide an effective mechanism for approximate Bayesian marginalization, and propose a related approach that further improves the predictive distribution by marginalizing within basins of attraction, without significant overhead. We"},"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":"2002.08791","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-20T15:13:27Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"de0c30862c39b5795637d1f3f91fddc007f750c216b225307c7b1b6b5f045584","abstract_canon_sha256":"6c58f7154f69176498df31f3ef18564176146616f905cf1afa324a1a04f5f519"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:09:54.263578Z","signature_b64":"HdIf7KyG8Goqtu+dZzO7+St0vYYfpcgZZ7zK84hMpjLb23cJNCHfmuPsj/IRaNxwpUThgQ9avrWnCMMl9pTCBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d1f1c22855ef85bd6ef0ce363629459ac0ca5265efcd373947e0830283f3b89","last_reissued_at":"2026-07-05T04:09:54.263085Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:09:54.263085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian Deep Learning and a Probabilistic Perspective of Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Gordon Wilson, Pavel Izmailov","submitted_at":"2020-02-20T15:13:27Z","abstract_excerpt":"The key distinguishing property of a Bayesian approach is marginalization, rather than using a single setting of weights. Bayesian marginalization can particularly improve the accuracy and calibration of modern deep neural networks, which are typically underspecified by the data, and can represent many compelling but different solutions. We show that deep ensembles provide an effective mechanism for approximate Bayesian marginalization, and propose a related approach that further improves the predictive distribution by marginalizing within basins of attraction, without significant overhead. We"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08791","kind":"arxiv","version":4},"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/2002.08791/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":"2002.08791","created_at":"2026-07-05T04:09:54.263141+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.08791v4","created_at":"2026-07-05T04:09:54.263141+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08791","created_at":"2026-07-05T04:09:54.263141+00:00"},{"alias_kind":"pith_short_12","alias_value":"JUPRYIUFL34F","created_at":"2026-07-05T04:09:54.263141+00:00"},{"alias_kind":"pith_short_16","alias_value":"JUPRYIUFL34FXVXP","created_at":"2026-07-05T04:09:54.263141+00:00"},{"alias_kind":"pith_short_8","alias_value":"JUPRYIUF","created_at":"2026-07-05T04:09:54.263141+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25000","citing_title":"Geo-Strat-RL: Learning Geological Event Reasoning from Verifiable Tasks","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21021","citing_title":"Continuous-Time Probabilistic Correctors for Uncertainty-Aware Physics-Based Spacecraft Trajectory Forecasting","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13262","citing_title":"Chem-GMNet: A Sphere-Native Geometric Transformer for Molecular Property Prediction","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05436","citing_title":"Estimating Implicit Regularization in Deep Learning","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG","json":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG.json","graph_json":"https://pith.science/api/pith-number/JUPRYIUFL34FXVXPBTRWGYUULG/graph.json","events_json":"https://pith.science/api/pith-number/JUPRYIUFL34FXVXPBTRWGYUULG/events.json","paper":"https://pith.science/paper/JUPRYIUF"},"agent_actions":{"view_html":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG","download_json":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG.json","view_paper":"https://pith.science/paper/JUPRYIUF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.08791&json=true","fetch_graph":"https://pith.science/api/pith-number/JUPRYIUFL34FXVXPBTRWGYUULG/graph.json","fetch_events":"https://pith.science/api/pith-number/JUPRYIUFL34FXVXPBTRWGYUULG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/action/storage_attestation","attest_author":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/action/author_attestation","sign_citation":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/action/citation_signature","submit_replication":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/action/replication_record"}},"created_at":"2026-07-05T04:09:54.263141+00:00","updated_at":"2026-07-05T04:09:54.263141+00:00"}