{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:ROPQWZ23OZ7TO23URXUZUWISLL","short_pith_number":"pith:ROPQWZ23","schema_version":"1.0","canonical_sha256":"8b9f0b675b767f376b748de99a59125aecbd38d4b806c4f38276ebe0ab12f35e","source":{"kind":"arxiv","id":"1902.01813","version":3},"attestation_state":"computed","paper":{"title":"Modular Block-diagonal Curvature Approximations for Feedforward Architectures","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Felix Dangel, Philipp Hennig, Stefan Harmeling","submitted_at":"2019-02-05T17:39:24Z","abstract_excerpt":"We propose a modular extension of backpropagation for the computation of block-diagonal approximations to various curvature matrices of the training objective (in particular, the Hessian, generalized Gauss-Newton, and positive-curvature Hessian). The approach reduces the otherwise tedious manual derivation of these matrices into local modules, and is easy to integrate into existing machine learning libraries. Moreover, we develop a compact notation derived from matrix differential calculus. We outline different strategies applicable to our method. They subsume recently-proposed block-diagonal "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1902.01813","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-05T17:39:24Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"5eadf9510bc295c2e480d0f53bc99eb3548f1aa7e405f11cde0d7b1300a7c2a3","abstract_canon_sha256":"f3286063112e84a099c395bd368ad5f92b28213b32c165b66275707ce5c93e4f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:44:24.762616Z","signature_b64":"BTylaHaFsOeVdeWmNdpq/pmpHNVfa5FTn6utqbP8vacmrNhv5NUQ23XWLVg0ND0n2oC1hgO+tjFw57fQTmPrAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b9f0b675b767f376b748de99a59125aecbd38d4b806c4f38276ebe0ab12f35e","last_reissued_at":"2026-07-05T00:44:24.762191Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:44:24.762191Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modular Block-diagonal Curvature Approximations for Feedforward Architectures","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Felix Dangel, Philipp Hennig, Stefan Harmeling","submitted_at":"2019-02-05T17:39:24Z","abstract_excerpt":"We propose a modular extension of backpropagation for the computation of block-diagonal approximations to various curvature matrices of the training objective (in particular, the Hessian, generalized Gauss-Newton, and positive-curvature Hessian). The approach reduces the otherwise tedious manual derivation of these matrices into local modules, and is easy to integrate into existing machine learning libraries. Moreover, we develop a compact notation derived from matrix differential calculus. We outline different strategies applicable to our method. They subsume recently-proposed block-diagonal "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.01813","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/1902.01813/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1902.01813","created_at":"2026-07-05T00:44:24.762256+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.01813v3","created_at":"2026-07-05T00:44:24.762256+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.01813","created_at":"2026-07-05T00:44:24.762256+00:00"},{"alias_kind":"pith_short_12","alias_value":"ROPQWZ23OZ7T","created_at":"2026-07-05T00:44:24.762256+00:00"},{"alias_kind":"pith_short_16","alias_value":"ROPQWZ23OZ7TO23U","created_at":"2026-07-05T00:44:24.762256+00:00"},{"alias_kind":"pith_short_8","alias_value":"ROPQWZ23","created_at":"2026-07-05T00:44:24.762256+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL","json":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL.json","graph_json":"https://pith.science/api/pith-number/ROPQWZ23OZ7TO23URXUZUWISLL/graph.json","events_json":"https://pith.science/api/pith-number/ROPQWZ23OZ7TO23URXUZUWISLL/events.json","paper":"https://pith.science/paper/ROPQWZ23"},"agent_actions":{"view_html":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL","download_json":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL.json","view_paper":"https://pith.science/paper/ROPQWZ23","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.01813&json=true","fetch_graph":"https://pith.science/api/pith-number/ROPQWZ23OZ7TO23URXUZUWISLL/graph.json","fetch_events":"https://pith.science/api/pith-number/ROPQWZ23OZ7TO23URXUZUWISLL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL/action/storage_attestation","attest_author":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL/action/author_attestation","sign_citation":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL/action/citation_signature","submit_replication":"https://pith.science/pith/ROPQWZ23OZ7TO23URXUZUWISLL/action/replication_record"}},"created_at":"2026-07-05T00:44:24.762256+00:00","updated_at":"2026-07-05T00:44:24.762256+00:00"}