{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:7NJ2FVC3733JRC32PEDUQM25GY","short_pith_number":"pith:7NJ2FVC3","schema_version":"1.0","canonical_sha256":"fb53a2d45bfef6988b7a790748335d363464e5298e9192b5ea278c64b692ad37","source":{"kind":"arxiv","id":"2307.13290","version":1},"attestation_state":"computed","paper":{"title":"Modify Training Directions in Function Space to Reduce Generalization Error","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Boyu Chen, Wenlian Lu, Yi Yu","submitted_at":"2023-07-25T07:11:30Z","abstract_excerpt":"We propose theoretical analyses of a modified natural gradient descent method in the neural network function space based on the eigendecompositions of neural tangent kernel and Fisher information matrix. We firstly present analytical expression for the function learned by this modified natural gradient under the assumptions of Gaussian distribution and infinite width limit. Thus, we explicitly derive the generalization error of the learned neural network function using theoretical methods from eigendecomposition and statistics theory. By decomposing of the total generalization error attributed"},"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":"2307.13290","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-07-25T07:11:30Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"fba2357739e6bf2aa2fad18af893a165638b7c62edb02f762067dfa0af76279f","abstract_canon_sha256":"db5765fcd691efc6b957db09c7d34a97d3e3918b135b5977f572a01a93c95066"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:34:11.911207Z","signature_b64":"SPZ0Ekb/iIs5LpIfR9yd/C+ef6VS4A5OuaxIeQSU3nlYpnaNB8No4yOYXOIXXbzl3Un1+/KwodZj25ynbe8YDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb53a2d45bfef6988b7a790748335d363464e5298e9192b5ea278c64b692ad37","last_reissued_at":"2026-07-05T06:34:11.910846Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:34:11.910846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modify Training Directions in Function Space to Reduce Generalization Error","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Boyu Chen, Wenlian Lu, Yi Yu","submitted_at":"2023-07-25T07:11:30Z","abstract_excerpt":"We propose theoretical analyses of a modified natural gradient descent method in the neural network function space based on the eigendecompositions of neural tangent kernel and Fisher information matrix. We firstly present analytical expression for the function learned by this modified natural gradient under the assumptions of Gaussian distribution and infinite width limit. Thus, we explicitly derive the generalization error of the learned neural network function using theoretical methods from eigendecomposition and statistics theory. By decomposing of the total generalization error attributed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.13290","kind":"arxiv","version":1},"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/2307.13290/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":"2307.13290","created_at":"2026-07-05T06:34:11.910901+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.13290v1","created_at":"2026-07-05T06:34:11.910901+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.13290","created_at":"2026-07-05T06:34:11.910901+00:00"},{"alias_kind":"pith_short_12","alias_value":"7NJ2FVC3733J","created_at":"2026-07-05T06:34:11.910901+00:00"},{"alias_kind":"pith_short_16","alias_value":"7NJ2FVC3733JRC32","created_at":"2026-07-05T06:34:11.910901+00:00"},{"alias_kind":"pith_short_8","alias_value":"7NJ2FVC3","created_at":"2026-07-05T06:34:11.910901+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/7NJ2FVC3733JRC32PEDUQM25GY","json":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY.json","graph_json":"https://pith.science/api/pith-number/7NJ2FVC3733JRC32PEDUQM25GY/graph.json","events_json":"https://pith.science/api/pith-number/7NJ2FVC3733JRC32PEDUQM25GY/events.json","paper":"https://pith.science/paper/7NJ2FVC3"},"agent_actions":{"view_html":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY","download_json":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY.json","view_paper":"https://pith.science/paper/7NJ2FVC3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.13290&json=true","fetch_graph":"https://pith.science/api/pith-number/7NJ2FVC3733JRC32PEDUQM25GY/graph.json","fetch_events":"https://pith.science/api/pith-number/7NJ2FVC3733JRC32PEDUQM25GY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY/action/storage_attestation","attest_author":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY/action/author_attestation","sign_citation":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY/action/citation_signature","submit_replication":"https://pith.science/pith/7NJ2FVC3733JRC32PEDUQM25GY/action/replication_record"}},"created_at":"2026-07-05T06:34:11.910901+00:00","updated_at":"2026-07-05T06:34:11.910901+00:00"}