{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:PDLUECTYBPSW56TVO32X45YDMI","short_pith_number":"pith:PDLUECTY","schema_version":"1.0","canonical_sha256":"78d7420a780be56efa7576f57e770362236f91f946ae062715fa5b2998aa42ee","source":{"kind":"arxiv","id":"2107.10884","version":3},"attestation_state":"computed","paper":{"title":"Structured second-order methods via natural gradient descent","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Frank Nielsen, Mark Schmidt, Mohammad Emtiyaz Khan, Wu Lin","submitted_at":"2021-07-22T19:03:53Z","abstract_excerpt":"In this paper, we propose new structured second-order methods and structured adaptive-gradient methods obtained by performing natural-gradient descent on structured parameter spaces. Natural-gradient descent is an attractive approach to design new algorithms in many settings such as gradient-free, adaptive-gradient, and second-order methods. Our structured methods not only enjoy a structural invariance but also admit a simple expression. Finally, we test the efficiency of our proposed methods on both deterministic non-convex problems and deep learning problems."},"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":"2107.10884","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"stat.ML","submitted_at":"2021-07-22T19:03:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5bd168fc5625701e2292192268a118822abcbf07facd96d859f0a168748ba7f8","abstract_canon_sha256":"1f60a92c2afa4618d82ae8ac8cc38c02664829b2de7d2d85b482274cab7fe7bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:58:16.755648Z","signature_b64":"MSdO1ro+tdCE6jKHvoCvsvOUf7ED0CxOpBzbAqMeOX6P1wk162hzkjqb/sKMO35MIXgB9wc+OsOuj7+ZfrcDAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78d7420a780be56efa7576f57e770362236f91f946ae062715fa5b2998aa42ee","last_reissued_at":"2026-07-05T03:58:16.755170Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:58:16.755170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Structured second-order methods via natural gradient descent","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Frank Nielsen, Mark Schmidt, Mohammad Emtiyaz Khan, Wu Lin","submitted_at":"2021-07-22T19:03:53Z","abstract_excerpt":"In this paper, we propose new structured second-order methods and structured adaptive-gradient methods obtained by performing natural-gradient descent on structured parameter spaces. Natural-gradient descent is an attractive approach to design new algorithms in many settings such as gradient-free, adaptive-gradient, and second-order methods. Our structured methods not only enjoy a structural invariance but also admit a simple expression. Finally, we test the efficiency of our proposed methods on both deterministic non-convex problems and deep learning problems."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.10884","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/2107.10884/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":"2107.10884","created_at":"2026-07-05T03:58:16.755226+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.10884v3","created_at":"2026-07-05T03:58:16.755226+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.10884","created_at":"2026-07-05T03:58:16.755226+00:00"},{"alias_kind":"pith_short_12","alias_value":"PDLUECTYBPSW","created_at":"2026-07-05T03:58:16.755226+00:00"},{"alias_kind":"pith_short_16","alias_value":"PDLUECTYBPSW56TV","created_at":"2026-07-05T03:58:16.755226+00:00"},{"alias_kind":"pith_short_8","alias_value":"PDLUECTY","created_at":"2026-07-05T03:58:16.755226+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.07853","citing_title":"Optimization Guarantees for Square-Root Natural-Gradient Variational Inference","ref_index":47,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI","json":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI.json","graph_json":"https://pith.science/api/pith-number/PDLUECTYBPSW56TVO32X45YDMI/graph.json","events_json":"https://pith.science/api/pith-number/PDLUECTYBPSW56TVO32X45YDMI/events.json","paper":"https://pith.science/paper/PDLUECTY"},"agent_actions":{"view_html":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI","download_json":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI.json","view_paper":"https://pith.science/paper/PDLUECTY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.10884&json=true","fetch_graph":"https://pith.science/api/pith-number/PDLUECTYBPSW56TVO32X45YDMI/graph.json","fetch_events":"https://pith.science/api/pith-number/PDLUECTYBPSW56TVO32X45YDMI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI/action/storage_attestation","attest_author":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI/action/author_attestation","sign_citation":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI/action/citation_signature","submit_replication":"https://pith.science/pith/PDLUECTYBPSW56TVO32X45YDMI/action/replication_record"}},"created_at":"2026-07-05T03:58:16.755226+00:00","updated_at":"2026-07-05T03:58:16.755226+00:00"}