{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:Q3J6KQVJ6HXXTPCLCND2V52N43","short_pith_number":"pith:Q3J6KQVJ","canonical_record":{"source":{"id":"1811.04504","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-11T23:18:27Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"0d73aa940a4016cea8cd7e4908882ace4d0ec42fa6ea5157be3826332c597497","abstract_canon_sha256":"bd00a5e3679fb5e470091692958eb6bf5daedc33e5b9cf6ddb87fa7a521c7bf0"},"schema_version":"1.0"},"canonical_sha256":"86d3e542a9f1ef79bc4b1347aaf74de6e804b373f277d8aa5395158608160744","source":{"kind":"arxiv","id":"1811.04504","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.04504","created_at":"2026-05-17T23:56:29Z"},{"alias_kind":"arxiv_version","alias_value":"1811.04504v2","created_at":"2026-05-17T23:56:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.04504","created_at":"2026-05-17T23:56:29Z"},{"alias_kind":"pith_short_12","alias_value":"Q3J6KQVJ6HXX","created_at":"2026-05-18T12:32:46Z"},{"alias_kind":"pith_short_16","alias_value":"Q3J6KQVJ6HXXTPCL","created_at":"2026-05-18T12:32:46Z"},{"alias_kind":"pith_short_8","alias_value":"Q3J6KQVJ","created_at":"2026-05-18T12:32:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:Q3J6KQVJ6HXXTPCLCND2V52N43","target":"record","payload":{"canonical_record":{"source":{"id":"1811.04504","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-11T23:18:27Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"0d73aa940a4016cea8cd7e4908882ace4d0ec42fa6ea5157be3826332c597497","abstract_canon_sha256":"bd00a5e3679fb5e470091692958eb6bf5daedc33e5b9cf6ddb87fa7a521c7bf0"},"schema_version":"1.0"},"canonical_sha256":"86d3e542a9f1ef79bc4b1347aaf74de6e804b373f277d8aa5395158608160744","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:56:29.214967Z","signature_b64":"VAs9RfcvwzYrw+pyuHyVsO9wvVKVvhrW1XHz/pgdZWpmz8UDhhwS8pENRzVi+4LB4k7lEl2jVNd+Gl8GGwjsAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86d3e542a9f1ef79bc4b1347aaf74de6e804b373f277d8aa5395158608160744","last_reissued_at":"2026-05-17T23:56:29.214485Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:56:29.214485Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.04504","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":"2026-05-17T23:56:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b2v5Lnh1/TxqH5oDeiRaV8oMb/2QxnMoP24ASn6wLlZo9OjZkfLNTIBAga2EG4EDOCvtPaNShFhOsdh51mXoAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T18:41:27.614309Z"},"content_sha256":"067a3d092d2a15fcb20d8170da003cf924fd6ba8d7e844c18cd2951b161fe296","schema_version":"1.0","event_id":"sha256:067a3d092d2a15fcb20d8170da003cf924fd6ba8d7e844c18cd2951b161fe296"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:Q3J6KQVJ6HXXTPCLCND2V52N43","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SLANG: Fast Structured Covariance Approximations for Bayesian Deep Learning with Natural Gradient","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Aaron Mishkin, Didrik Nielsen, Frederik Kunstner, Mark Schmidt, Mohammad Emtiyaz Khan","submitted_at":"2018-11-11T23:18:27Z","abstract_excerpt":"Uncertainty estimation in large deep-learning models is a computationally challenging task, where it is difficult to form even a Gaussian approximation to the posterior distribution. In such situations, existing methods usually resort to a diagonal approximation of the covariance matrix despite, the fact that these matrices are known to result in poor uncertainty estimates. To address this issue, we propose a new stochastic, low-rank, approximate natural-gradient (SLANG) method for variational inference in large, deep models. Our method estimates a \"diagonal plus low-rank\" structure based sole"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.04504","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":"2026-05-17T23:56:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hyFCU2WLSTDNGlidRZgbpWgBHagPHVEmwKVKSiZHpZdlMbqABkgx6ehP/d1VRdbp3j1O/nviePKPxVERwBngCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T18:41:27.615012Z"},"content_sha256":"7e8eb4e1a4eec98c131b4d027575dd466ba047832127cd08fbb090496509f743","schema_version":"1.0","event_id":"sha256:7e8eb4e1a4eec98c131b4d027575dd466ba047832127cd08fbb090496509f743"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q3J6KQVJ6HXXTPCLCND2V52N43/bundle.json","state_url":"https://pith.science/pith/Q3J6KQVJ6HXXTPCLCND2V52N43/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q3J6KQVJ6HXXTPCLCND2V52N43/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":"2026-05-28T18:41:27Z","links":{"resolver":"https://pith.science/pith/Q3J6KQVJ6HXXTPCLCND2V52N43","bundle":"https://pith.science/pith/Q3J6KQVJ6HXXTPCLCND2V52N43/bundle.json","state":"https://pith.science/pith/Q3J6KQVJ6HXXTPCLCND2V52N43/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q3J6KQVJ6HXXTPCLCND2V52N43/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:Q3J6KQVJ6HXXTPCLCND2V52N43","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":"bd00a5e3679fb5e470091692958eb6bf5daedc33e5b9cf6ddb87fa7a521c7bf0","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-11T23:18:27Z","title_canon_sha256":"0d73aa940a4016cea8cd7e4908882ace4d0ec42fa6ea5157be3826332c597497"},"schema_version":"1.0","source":{"id":"1811.04504","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.04504","created_at":"2026-05-17T23:56:29Z"},{"alias_kind":"arxiv_version","alias_value":"1811.04504v2","created_at":"2026-05-17T23:56:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.04504","created_at":"2026-05-17T23:56:29Z"},{"alias_kind":"pith_short_12","alias_value":"Q3J6KQVJ6HXX","created_at":"2026-05-18T12:32:46Z"},{"alias_kind":"pith_short_16","alias_value":"Q3J6KQVJ6HXXTPCL","created_at":"2026-05-18T12:32:46Z"},{"alias_kind":"pith_short_8","alias_value":"Q3J6KQVJ","created_at":"2026-05-18T12:32:46Z"}],"graph_snapshots":[{"event_id":"sha256:7e8eb4e1a4eec98c131b4d027575dd466ba047832127cd08fbb090496509f743","target":"graph","created_at":"2026-05-17T23:56:29Z","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":"Uncertainty estimation in large deep-learning models is a computationally challenging task, where it is difficult to form even a Gaussian approximation to the posterior distribution. In such situations, existing methods usually resort to a diagonal approximation of the covariance matrix despite, the fact that these matrices are known to result in poor uncertainty estimates. To address this issue, we propose a new stochastic, low-rank, approximate natural-gradient (SLANG) method for variational inference in large, deep models. Our method estimates a \"diagonal plus low-rank\" structure based sole","authors_text":"Aaron Mishkin, Didrik Nielsen, Frederik Kunstner, Mark Schmidt, Mohammad Emtiyaz Khan","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-11T23:18:27Z","title":"SLANG: Fast Structured Covariance Approximations for Bayesian Deep Learning with Natural Gradient"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.04504","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":"sha256:067a3d092d2a15fcb20d8170da003cf924fd6ba8d7e844c18cd2951b161fe296","target":"record","created_at":"2026-05-17T23:56:29Z","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":"bd00a5e3679fb5e470091692958eb6bf5daedc33e5b9cf6ddb87fa7a521c7bf0","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-11T23:18:27Z","title_canon_sha256":"0d73aa940a4016cea8cd7e4908882ace4d0ec42fa6ea5157be3826332c597497"},"schema_version":"1.0","source":{"id":"1811.04504","kind":"arxiv","version":2}},"canonical_sha256":"86d3e542a9f1ef79bc4b1347aaf74de6e804b373f277d8aa5395158608160744","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"86d3e542a9f1ef79bc4b1347aaf74de6e804b373f277d8aa5395158608160744","first_computed_at":"2026-05-17T23:56:29.214485Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:56:29.214485Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VAs9RfcvwzYrw+pyuHyVsO9wvVKVvhrW1XHz/pgdZWpmz8UDhhwS8pENRzVi+4LB4k7lEl2jVNd+Gl8GGwjsAA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:56:29.214967Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.04504","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:067a3d092d2a15fcb20d8170da003cf924fd6ba8d7e844c18cd2951b161fe296","sha256:7e8eb4e1a4eec98c131b4d027575dd466ba047832127cd08fbb090496509f743"],"state_sha256":"817da12e276959a62a58160ac0431e870c701484e4217774a8ada90caf505c5c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BZuj5XFIKuL4toVUZpyGUN+WCnUgjQ9sI9MBQqvobcbyj/KiUlpc0mxSGwGY9/uFg5FG5lel+V/s78g7c4vSAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T18:41:27.618580Z","bundle_sha256":"7a37e5c4dc4bf28df5c1b1f072818d3b08b0594a5be6e4f549c2389f7c0bbc51"}}