{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:ZUVZ4QCKYARHV5DJOS7KNAPC2P","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":"22f86f1ef82b45b418d4f81c32c7e098f69618fb16006c3deac89046d4ee9d2c","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-05T23:13:42Z","title_canon_sha256":"546549736edb6b8b7211035dd6a9215efbe3ba485c963cf1214e7c874f67a43c"},"schema_version":"1.0","source":{"id":"1906.02353","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.02353","created_at":"2026-05-17T23:44:01Z"},{"alias_kind":"arxiv_version","alias_value":"1906.02353v1","created_at":"2026-05-17T23:44:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.02353","created_at":"2026-05-17T23:44:01Z"},{"alias_kind":"pith_short_12","alias_value":"ZUVZ4QCKYARH","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"ZUVZ4QCKYARHV5DJ","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"ZUVZ4QCK","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:dd79cd2212ac1c554a2d219fa07d36fd96e4f0842888179027b49520f9b523f1","target":"graph","created_at":"2026-05-17T23:44:01Z","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 present practical Levenberg-Marquardt variants of Gauss-Newton and natural gradient methods for solving non-convex optimization problems that arise in training deep neural networks involving enormous numbers of variables and huge data sets. Our methods use subsampled Gauss-Newton or Fisher information matrices and either subsampled gradient estimates (fully stochastic) or full gradients (semi-stochastic), which, in the latter case, we prove convergent to a stationary point. By using the Sherman-Morrison-Woodbury formula with automatic differentiation (backpropagation) we show how our method","authors_text":"Donald Goldfarb, Yi Ren","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-05T23:13:42Z","title":"Efficient Subsampled Gauss-Newton and Natural Gradient Methods for Training Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.02353","kind":"arxiv","version":1},"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":"sha256:727a0ed18f96c5d79c2c03aff78ce48c6f3e862138f05bef6f5e5a2c3fae091d","target":"record","created_at":"2026-05-17T23:44:01Z","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":"22f86f1ef82b45b418d4f81c32c7e098f69618fb16006c3deac89046d4ee9d2c","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-05T23:13:42Z","title_canon_sha256":"546549736edb6b8b7211035dd6a9215efbe3ba485c963cf1214e7c874f67a43c"},"schema_version":"1.0","source":{"id":"1906.02353","kind":"arxiv","version":1}},"canonical_sha256":"cd2b9e404ac0227af46974bea681e2d3d39e4fb5c19a006ffe6312b81193930c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cd2b9e404ac0227af46974bea681e2d3d39e4fb5c19a006ffe6312b81193930c","first_computed_at":"2026-05-17T23:44:01.491978Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:44:01.491978Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KfLkA9wv1TquaTSOz8zl5A+67L3bSkZLXNVVyNYQhBvt30HDM9m9HpnA1a/z/g3CQrDeta6QCNOo7OLMsdgVDQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:44:01.492461Z","signed_message":"canonical_sha256_bytes"},"source_id":"1906.02353","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:727a0ed18f96c5d79c2c03aff78ce48c6f3e862138f05bef6f5e5a2c3fae091d","sha256:dd79cd2212ac1c554a2d219fa07d36fd96e4f0842888179027b49520f9b523f1"],"state_sha256":"7a82977bf07108851e809ea4ffe70510a113e97e00a45d87f1e664ea90da94fa"}