{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IEBWWZMW662P2ZRKKAUM7M4P4Z","short_pith_number":"pith:IEBWWZMW","schema_version":"1.0","canonical_sha256":"41036b6596f7b4fd662a5028cfb38fe657937efd998810a1b43073fa53fb326f","source":{"kind":"arxiv","id":"2605.18180","version":1},"attestation_state":"computed","paper":{"title":"Canonical Regularisation of Wide Feature-Learning Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"George Whittle, Juliusz Ziomek, Maike A. Osborne, Natalia Ares, Pranav Vaidhyanathan","submitted_at":"2026-05-18T10:23:06Z","abstract_excerpt":"Wide neural networks in the feature-learning regime drive modern deep learning, and yet they remain far less studied than their kernel-regime counterparts. We consider a critical yet under-explored difference between these two regimes: the regulariser and prior implied by gradient flow training. This canonical regularisation property is well-studied in kernel regime networks -- of all the infinite global minima, gradient flow selects exactly the vanishing ridge solution -- and underpins the celebrated NN-GP correspondence, precisely allowing the modelling of noise during training. However, we "},"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":"2605.18180","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-18T10:23:06Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e84cf96b6e41fe6718a80604036a66859e6cb47014f1ba759474190ad96ee5df","abstract_canon_sha256":"e2e9732572379973385baf16aeaadd6e2af9ffc2587ddc2326c0137e85f80e3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:49.533788Z","signature_b64":"HhrtdrwlJ/RAi8ILJ+gzetljiXgy9E2oZi8JV77jCkIDGcSX45dp1A0Q84h05fiVP35jP7yDmhq7xisMhi4gBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41036b6596f7b4fd662a5028cfb38fe657937efd998810a1b43073fa53fb326f","last_reissued_at":"2026-05-20T00:05:49.533170Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:49.533170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Canonical Regularisation of Wide Feature-Learning Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"George Whittle, Juliusz Ziomek, Maike A. Osborne, Natalia Ares, Pranav Vaidhyanathan","submitted_at":"2026-05-18T10:23:06Z","abstract_excerpt":"Wide neural networks in the feature-learning regime drive modern deep learning, and yet they remain far less studied than their kernel-regime counterparts. We consider a critical yet under-explored difference between these two regimes: the regulariser and prior implied by gradient flow training. This canonical regularisation property is well-studied in kernel regime networks -- of all the infinite global minima, gradient flow selects exactly the vanishing ridge solution -- and underpins the celebrated NN-GP correspondence, precisely allowing the modelling of noise during training. However, we "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18180","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/2605.18180/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-19T23:41:59.030110Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T23:33:35.343507Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"70b0e9d211df93c6fce0ca2864d268d35e006b29c9224661168a935e1815225c"},"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":"2605.18180","created_at":"2026-05-20T00:05:49.533250+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18180v1","created_at":"2026-05-20T00:05:49.533250+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18180","created_at":"2026-05-20T00:05:49.533250+00:00"},{"alias_kind":"pith_short_12","alias_value":"IEBWWZMW662P","created_at":"2026-05-20T00:05:49.533250+00:00"},{"alias_kind":"pith_short_16","alias_value":"IEBWWZMW662P2ZRK","created_at":"2026-05-20T00:05:49.533250+00:00"},{"alias_kind":"pith_short_8","alias_value":"IEBWWZMW","created_at":"2026-05-20T00:05:49.533250+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/IEBWWZMW662P2ZRKKAUM7M4P4Z","json":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z.json","graph_json":"https://pith.science/api/pith-number/IEBWWZMW662P2ZRKKAUM7M4P4Z/graph.json","events_json":"https://pith.science/api/pith-number/IEBWWZMW662P2ZRKKAUM7M4P4Z/events.json","paper":"https://pith.science/paper/IEBWWZMW"},"agent_actions":{"view_html":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z","download_json":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z.json","view_paper":"https://pith.science/paper/IEBWWZMW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18180&json=true","fetch_graph":"https://pith.science/api/pith-number/IEBWWZMW662P2ZRKKAUM7M4P4Z/graph.json","fetch_events":"https://pith.science/api/pith-number/IEBWWZMW662P2ZRKKAUM7M4P4Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z/action/storage_attestation","attest_author":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z/action/author_attestation","sign_citation":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z/action/citation_signature","submit_replication":"https://pith.science/pith/IEBWWZMW662P2ZRKKAUM7M4P4Z/action/replication_record"}},"created_at":"2026-05-20T00:05:49.533250+00:00","updated_at":"2026-05-20T00:05:49.533250+00:00"}