{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:HJW3LU3VJA3KHWJLQ7MDBAQQEE","short_pith_number":"pith:HJW3LU3V","canonical_record":{"source":{"id":"2008.08400","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-08-19T12:35:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6b7cc14382c9d8a0b1c428e8da8c619df32fd457884f0ffb3509e88bb55aae34","abstract_canon_sha256":"d2c86f1e70ddf6ba16ba10c6ed1c299d1fb86e31938f834fa3fd6216b63a7b4c"},"schema_version":"1.0"},"canonical_sha256":"3a6db5d3754836a3d92b87d8308210210e3d7d7a5e28d3f2313bf02e11fb1e32","source":{"kind":"arxiv","id":"2008.08400","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.08400","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"arxiv_version","alias_value":"2008.08400v3","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.08400","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"pith_short_12","alias_value":"HJW3LU3VJA3K","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"pith_short_16","alias_value":"HJW3LU3VJA3KHWJL","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"pith_short_8","alias_value":"HJW3LU3V","created_at":"2026-07-05T02:18:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:HJW3LU3VJA3KHWJLQ7MDBAQQEE","target":"record","payload":{"canonical_record":{"source":{"id":"2008.08400","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-08-19T12:35:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6b7cc14382c9d8a0b1c428e8da8c619df32fd457884f0ffb3509e88bb55aae34","abstract_canon_sha256":"d2c86f1e70ddf6ba16ba10c6ed1c299d1fb86e31938f834fa3fd6216b63a7b4c"},"schema_version":"1.0"},"canonical_sha256":"3a6db5d3754836a3d92b87d8308210210e3d7d7a5e28d3f2313bf02e11fb1e32","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:18:17.412106Z","signature_b64":"3YovH1UhlfWNhG2f9ZqRpTpIdLTbrjRkIEPxuSr1SNTHo0MEwtzIT+xGZ4nkMiOLyh77lHpOvppd9B+ZgEbMBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a6db5d3754836a3d92b87d8308210210e3d7d7a5e28d3f2313bf02e11fb1e32","last_reissued_at":"2026-07-05T02:18:17.411730Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:18:17.411730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2008.08400","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:18:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UifqFeSlROLzXT0O61xy4DBbNOv7aHTv2k9BXUy9qZpvnJsE7vNZJr04EVAsmgtaktD0np8pt4GT+z/JUOggDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T06:57:17.648880Z"},"content_sha256":"d4d2008264b80dad6bb55686b7b3d99eda60bfddf825ff981bb562076210ebf2","schema_version":"1.0","event_id":"sha256:d4d2008264b80dad6bb55686b7b3d99eda60bfddf825ff981bb562076210ebf2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:HJW3LU3VJA3KHWJLQ7MDBAQQEE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving predictions of Bayesian neural nets via local linearization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Immer, Maciej Korzepa, Matthias Bauer","submitted_at":"2020-08-19T12:35:55Z","abstract_excerpt":"The generalized Gauss-Newton (GGN) approximation is often used to make practical Bayesian deep learning approaches scalable by replacing a second order derivative with a product of first order derivatives. In this paper we argue that the GGN approximation should be understood as a local linearization of the underlying Bayesian neural network (BNN), which turns the BNN into a generalized linear model (GLM). Because we use this linearized model for posterior inference, we should also predict using this modified model instead of the original one. We refer to this modified predictive as \"GLM predi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.08400","kind":"arxiv","version":3},"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/2008.08400/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":"2026-07-05T02:18:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EUBkbCaVsfHO67aaoCQaINHQG/+wbLJzjzX1EEPn225ZnGUkw+uJ4mHxKU0AgyJ7o5JOnqMPo9370dd5lU/jDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T06:57:17.649368Z"},"content_sha256":"1a2a5925e418b1a4e2cb31e49ccf9623ea13185dd6fd5465613a2005792f0da4","schema_version":"1.0","event_id":"sha256:1a2a5925e418b1a4e2cb31e49ccf9623ea13185dd6fd5465613a2005792f0da4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE/bundle.json","state_url":"https://pith.science/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE/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":"2026-08-04T06:57:17Z","links":{"resolver":"https://pith.science/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE","bundle":"https://pith.science/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE/bundle.json","state":"https://pith.science/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HJW3LU3VJA3KHWJLQ7MDBAQQEE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:HJW3LU3VJA3KHWJLQ7MDBAQQEE","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":"d2c86f1e70ddf6ba16ba10c6ed1c299d1fb86e31938f834fa3fd6216b63a7b4c","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-08-19T12:35:55Z","title_canon_sha256":"6b7cc14382c9d8a0b1c428e8da8c619df32fd457884f0ffb3509e88bb55aae34"},"schema_version":"1.0","source":{"id":"2008.08400","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.08400","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"arxiv_version","alias_value":"2008.08400v3","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.08400","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"pith_short_12","alias_value":"HJW3LU3VJA3K","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"pith_short_16","alias_value":"HJW3LU3VJA3KHWJL","created_at":"2026-07-05T02:18:17Z"},{"alias_kind":"pith_short_8","alias_value":"HJW3LU3V","created_at":"2026-07-05T02:18:17Z"}],"graph_snapshots":[{"event_id":"sha256:1a2a5925e418b1a4e2cb31e49ccf9623ea13185dd6fd5465613a2005792f0da4","target":"graph","created_at":"2026-07-05T02:18:17Z","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/2008.08400/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The generalized Gauss-Newton (GGN) approximation is often used to make practical Bayesian deep learning approaches scalable by replacing a second order derivative with a product of first order derivatives. In this paper we argue that the GGN approximation should be understood as a local linearization of the underlying Bayesian neural network (BNN), which turns the BNN into a generalized linear model (GLM). Because we use this linearized model for posterior inference, we should also predict using this modified model instead of the original one. We refer to this modified predictive as \"GLM predi","authors_text":"Alexander Immer, Maciej Korzepa, Matthias Bauer","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-08-19T12:35:55Z","title":"Improving predictions of Bayesian neural nets via local linearization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.08400","kind":"arxiv","version":3},"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:d4d2008264b80dad6bb55686b7b3d99eda60bfddf825ff981bb562076210ebf2","target":"record","created_at":"2026-07-05T02:18:17Z","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":"d2c86f1e70ddf6ba16ba10c6ed1c299d1fb86e31938f834fa3fd6216b63a7b4c","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-08-19T12:35:55Z","title_canon_sha256":"6b7cc14382c9d8a0b1c428e8da8c619df32fd457884f0ffb3509e88bb55aae34"},"schema_version":"1.0","source":{"id":"2008.08400","kind":"arxiv","version":3}},"canonical_sha256":"3a6db5d3754836a3d92b87d8308210210e3d7d7a5e28d3f2313bf02e11fb1e32","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3a6db5d3754836a3d92b87d8308210210e3d7d7a5e28d3f2313bf02e11fb1e32","first_computed_at":"2026-07-05T02:18:17.411730Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:18:17.411730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3YovH1UhlfWNhG2f9ZqRpTpIdLTbrjRkIEPxuSr1SNTHo0MEwtzIT+xGZ4nkMiOLyh77lHpOvppd9B+ZgEbMBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:18:17.412106Z","signed_message":"canonical_sha256_bytes"},"source_id":"2008.08400","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4d2008264b80dad6bb55686b7b3d99eda60bfddf825ff981bb562076210ebf2","sha256:1a2a5925e418b1a4e2cb31e49ccf9623ea13185dd6fd5465613a2005792f0da4"],"state_sha256":"9765820fb22d1866237ac336b57640535e7e41e511c7c8b830300b83907302c0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nBwWh3nAH00fX2OLGjaVShBpiHLG4lw69vBdjIJjxPAsLWLKuIrFq/pEJ4cLiXBEomWNOV8nYVQCBb+3iOceBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T06:57:17.653070Z","bundle_sha256":"d993c52bca16b124617bd5076ff88013e8b66eb2550c575f1525454bdd7cc470"}}