{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:JUPRYIUFL34FXVXPBTRWGYUULG","short_pith_number":"pith:JUPRYIUF","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"},"canonical_sha256":"4d1f1c22855ef85bd6ef0ce363629459ac0ca5265efcd373947e0830283f3b89","source":{"kind":"arxiv","id":"2002.08791","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2002.08791","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"arxiv_version","alias_value":"2002.08791v4","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08791","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"pith_short_12","alias_value":"JUPRYIUFL34F","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"pith_short_16","alias_value":"JUPRYIUFL34FXVXP","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"pith_short_8","alias_value":"JUPRYIUF","created_at":"2026-07-05T04:09:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:JUPRYIUFL34FXVXPBTRWGYUULG","target":"record","payload":{"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"},"canonical_sha256":"4d1f1c22855ef85bd6ef0ce363629459ac0ca5265efcd373947e0830283f3b89","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"},"source_kind":"arxiv","source_id":"2002.08791","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:09:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q08dO1wpHxm4Y+Z9Qv7pCnHt2Y4sDL7ZCAVgZInM8zIPK4TFCoSG0BucL9LOflJ5EYwNdry4TEi6Ta461utNCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T01:22:57.535460Z"},"content_sha256":"72e10aa4b7a60627bc7ad51dd8881a3d20b913b01606f7742ee6e71a95ec99f4","schema_version":"1.0","event_id":"sha256:72e10aa4b7a60627bc7ad51dd8881a3d20b913b01606f7742ee6e71a95ec99f4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:JUPRYIUFL34FXVXPBTRWGYUULG","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:09:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/h7/piBA5C5TwZfCUsVx+4v31p9Gr8IqOSzeF2HRRu7uOxgNrFQTcFIVaxUBUr9kGL3cD9qswfBUmUxvXhjlCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T01:22:57.535986Z"},"content_sha256":"7eb389ca7743b04a175040c65c12b7b2182002258b38589fa81899892f0377bf","schema_version":"1.0","event_id":"sha256:7eb389ca7743b04a175040c65c12b7b2182002258b38589fa81899892f0377bf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/bundle.json","state_url":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JUPRYIUFL34FXVXPBTRWGYUULG/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-10T01:22:57Z","links":{"resolver":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG","bundle":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/bundle.json","state":"https://pith.science/pith/JUPRYIUFL34FXVXPBTRWGYUULG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JUPRYIUFL34FXVXPBTRWGYUULG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:JUPRYIUFL34FXVXPBTRWGYUULG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6c58f7154f69176498df31f3ef18564176146616f905cf1afa324a1a04f5f519","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-20T15:13:27Z","title_canon_sha256":"de0c30862c39b5795637d1f3f91fddc007f750c216b225307c7b1b6b5f045584"},"schema_version":"1.0","source":{"id":"2002.08791","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2002.08791","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"arxiv_version","alias_value":"2002.08791v4","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08791","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"pith_short_12","alias_value":"JUPRYIUFL34F","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"pith_short_16","alias_value":"JUPRYIUFL34FXVXP","created_at":"2026-07-05T04:09:54Z"},{"alias_kind":"pith_short_8","alias_value":"JUPRYIUF","created_at":"2026-07-05T04:09:54Z"}],"graph_snapshots":[{"event_id":"sha256:7eb389ca7743b04a175040c65c12b7b2182002258b38589fa81899892f0377bf","target":"graph","created_at":"2026-07-05T04:09:54Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2002.08791/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Andrew Gordon Wilson, Pavel Izmailov","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-20T15:13:27Z","title":"Bayesian Deep Learning and a Probabilistic Perspective of Generalization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08791","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:72e10aa4b7a60627bc7ad51dd8881a3d20b913b01606f7742ee6e71a95ec99f4","target":"record","created_at":"2026-07-05T04:09:54Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6c58f7154f69176498df31f3ef18564176146616f905cf1afa324a1a04f5f519","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-20T15:13:27Z","title_canon_sha256":"de0c30862c39b5795637d1f3f91fddc007f750c216b225307c7b1b6b5f045584"},"schema_version":"1.0","source":{"id":"2002.08791","kind":"arxiv","version":4}},"canonical_sha256":"4d1f1c22855ef85bd6ef0ce363629459ac0ca5265efcd373947e0830283f3b89","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4d1f1c22855ef85bd6ef0ce363629459ac0ca5265efcd373947e0830283f3b89","first_computed_at":"2026-07-05T04:09:54.263085Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:09:54.263085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HdIf7KyG8Goqtu+dZzO7+St0vYYfpcgZZ7zK84hMpjLb23cJNCHfmuPsj/IRaNxwpUThgQ9avrWnCMMl9pTCBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T04:09:54.263578Z","signed_message":"canonical_sha256_bytes"},"source_id":"2002.08791","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:72e10aa4b7a60627bc7ad51dd8881a3d20b913b01606f7742ee6e71a95ec99f4","sha256:7eb389ca7743b04a175040c65c12b7b2182002258b38589fa81899892f0377bf"],"state_sha256":"666dd3f0bb5f4affe0153890a21066447c31efb0ab2ebc648a340fb363eda5d4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lkNhvl781BcIWwHKAINMcBqpMsnbGRkahcbObrFkCNaOenv20futlip37bILBjr4+pHUi14WGpIHCm1t845YDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T01:22:57.540611Z","bundle_sha256":"61ae8c1c253a541512650fde7c123f9a8b119e442062abaae3fbac2f6220e0e2"}}