{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:PYXUHHXS6XKLTZGVEUSNCC27OA","short_pith_number":"pith:PYXUHHXS","canonical_record":{"source":{"id":"1703.01961","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-03-06T16:39:16Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8bf22921e72bbebc3b7dcd9324326991ee01135dd896fa76b214ca4850fbd626","abstract_canon_sha256":"7d863a30c801349ce54abba79ad0b4016ff1db66a79078e28d42144616ace97b"},"schema_version":"1.0"},"canonical_sha256":"7e2f439ef2f5d4b9e4d52524d10b5f700d96a86b105e8c4473513052e54a6d3c","source":{"kind":"arxiv","id":"1703.01961","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1703.01961","created_at":"2026-05-18T00:42:29Z"},{"alias_kind":"arxiv_version","alias_value":"1703.01961v2","created_at":"2026-05-18T00:42:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.01961","created_at":"2026-05-18T00:42:29Z"},{"alias_kind":"pith_short_12","alias_value":"PYXUHHXS6XKL","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"PYXUHHXS6XKLTZGV","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"PYXUHHXS","created_at":"2026-05-18T12:31:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:PYXUHHXS6XKLTZGVEUSNCC27OA","target":"record","payload":{"canonical_record":{"source":{"id":"1703.01961","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-03-06T16:39:16Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8bf22921e72bbebc3b7dcd9324326991ee01135dd896fa76b214ca4850fbd626","abstract_canon_sha256":"7d863a30c801349ce54abba79ad0b4016ff1db66a79078e28d42144616ace97b"},"schema_version":"1.0"},"canonical_sha256":"7e2f439ef2f5d4b9e4d52524d10b5f700d96a86b105e8c4473513052e54a6d3c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:42:29.449268Z","signature_b64":"t3hkJewmtuVFGr9JNYZiI3FWMsVfd6KvCk7FCeZZzBKGGQ6Gttoy/7N5z0UCmVkVacV+qbLXs162EBcwV23aAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e2f439ef2f5d4b9e4d52524d10b5f700d96a86b105e8c4473513052e54a6d3c","last_reissued_at":"2026-05-18T00:42:29.448640Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:42:29.448640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1703.01961","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:42:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yelid/jaQIwV2fe4MAiaOAL+IR30tC6xbZjpgYEyRkDUVnh4WBmc7BdGc6OjygxPaCucBuYqbZ7982oQd0FBBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T09:11:24.182149Z"},"content_sha256":"e8a6de6df74d5538259884f7e043b7ece198914dbeebc619a98523f8fa996620","schema_version":"1.0","event_id":"sha256:e8a6de6df74d5538259884f7e043b7ece198914dbeebc619a98523f8fa996620"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:PYXUHHXS6XKLTZGVEUSNCC27OA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multiplicative Normalizing Flows for Variational Bayesian Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Christos Louizos, Max Welling","submitted_at":"2017-03-06T16:39:16Z","abstract_excerpt":"We reinterpret multiplicative noise in neural networks as auxiliary random variables that augment the approximate posterior in a variational setting for Bayesian neural networks. We show that through this interpretation it is both efficient and straightforward to improve the approximation by employing normalizing flows while still allowing for local reparametrizations and a tractable lower bound. In experiments we show that with this new approximation we can significantly improve upon classical mean field for Bayesian neural networks on both predictive accuracy as well as predictive uncertaint"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.01961","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2026-05-18T00:42:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T3msvKuAh3RCaf9VY7AhKdzvudzvVU5kzATI4scUqOXYWAH3TneMgi6qccy32t68vYRzOpoGlxUw6kJpkRNoDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T09:11:24.182894Z"},"content_sha256":"3b8c9bfd53104c42fc89a8175832b121065b54da46d4e61c28aee17ca3fb3340","schema_version":"1.0","event_id":"sha256:3b8c9bfd53104c42fc89a8175832b121065b54da46d4e61c28aee17ca3fb3340"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PYXUHHXS6XKLTZGVEUSNCC27OA/bundle.json","state_url":"https://pith.science/pith/PYXUHHXS6XKLTZGVEUSNCC27OA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PYXUHHXS6XKLTZGVEUSNCC27OA/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":"2026-05-27T09:11:24Z","links":{"resolver":"https://pith.science/pith/PYXUHHXS6XKLTZGVEUSNCC27OA","bundle":"https://pith.science/pith/PYXUHHXS6XKLTZGVEUSNCC27OA/bundle.json","state":"https://pith.science/pith/PYXUHHXS6XKLTZGVEUSNCC27OA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PYXUHHXS6XKLTZGVEUSNCC27OA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:PYXUHHXS6XKLTZGVEUSNCC27OA","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":"7d863a30c801349ce54abba79ad0b4016ff1db66a79078e28d42144616ace97b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-03-06T16:39:16Z","title_canon_sha256":"8bf22921e72bbebc3b7dcd9324326991ee01135dd896fa76b214ca4850fbd626"},"schema_version":"1.0","source":{"id":"1703.01961","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1703.01961","created_at":"2026-05-18T00:42:29Z"},{"alias_kind":"arxiv_version","alias_value":"1703.01961v2","created_at":"2026-05-18T00:42:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.01961","created_at":"2026-05-18T00:42:29Z"},{"alias_kind":"pith_short_12","alias_value":"PYXUHHXS6XKL","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"PYXUHHXS6XKLTZGV","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"PYXUHHXS","created_at":"2026-05-18T12:31:37Z"}],"graph_snapshots":[{"event_id":"sha256:3b8c9bfd53104c42fc89a8175832b121065b54da46d4e61c28aee17ca3fb3340","target":"graph","created_at":"2026-05-18T00:42:29Z","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"},"paper":{"abstract_excerpt":"We reinterpret multiplicative noise in neural networks as auxiliary random variables that augment the approximate posterior in a variational setting for Bayesian neural networks. We show that through this interpretation it is both efficient and straightforward to improve the approximation by employing normalizing flows while still allowing for local reparametrizations and a tractable lower bound. In experiments we show that with this new approximation we can significantly improve upon classical mean field for Bayesian neural networks on both predictive accuracy as well as predictive uncertaint","authors_text":"Christos Louizos, Max Welling","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-03-06T16:39:16Z","title":"Multiplicative Normalizing Flows for Variational Bayesian Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.01961","kind":"arxiv","version":2},"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:e8a6de6df74d5538259884f7e043b7ece198914dbeebc619a98523f8fa996620","target":"record","created_at":"2026-05-18T00:42:29Z","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":"7d863a30c801349ce54abba79ad0b4016ff1db66a79078e28d42144616ace97b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-03-06T16:39:16Z","title_canon_sha256":"8bf22921e72bbebc3b7dcd9324326991ee01135dd896fa76b214ca4850fbd626"},"schema_version":"1.0","source":{"id":"1703.01961","kind":"arxiv","version":2}},"canonical_sha256":"7e2f439ef2f5d4b9e4d52524d10b5f700d96a86b105e8c4473513052e54a6d3c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7e2f439ef2f5d4b9e4d52524d10b5f700d96a86b105e8c4473513052e54a6d3c","first_computed_at":"2026-05-18T00:42:29.448640Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:42:29.448640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"t3hkJewmtuVFGr9JNYZiI3FWMsVfd6KvCk7FCeZZzBKGGQ6Gttoy/7N5z0UCmVkVacV+qbLXs162EBcwV23aAg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:42:29.449268Z","signed_message":"canonical_sha256_bytes"},"source_id":"1703.01961","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e8a6de6df74d5538259884f7e043b7ece198914dbeebc619a98523f8fa996620","sha256:3b8c9bfd53104c42fc89a8175832b121065b54da46d4e61c28aee17ca3fb3340"],"state_sha256":"d3bc62d34a709b753e5a219a43572ded98b5d1e27fa470e392095b881c3a1c9c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XvWg32XGzjXSG2ldMgNNb63pkQF3aziAYfo7K6VdGL+/PCnfira3S38a5byRprx9jYH2d6tzJ4gXGqp1HxmdDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T09:11:24.186825Z","bundle_sha256":"ae471658c32a2ad6f131767f7e18f495725e9eb76b6ee56d0f86dbe744c0883a"}}