{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:MGWIPGPKJ252ZZ44A3FLHFQTLS","short_pith_number":"pith:MGWIPGPK","schema_version":"1.0","canonical_sha256":"61ac8799ea4ebbace79c06cab396135c80ce60f39bce51231fdc9f4f4bad0f4b","source":{"kind":"arxiv","id":"2309.01213","version":3},"attestation_state":"computed","paper":{"title":"Implicit regularization of deep residual networks towards neural ODEs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"G\\'erard Biau, Michael E. Sander, Pierre Marion, Yu-Han Wu","submitted_at":"2023-09-03T16:35:59Z","abstract_excerpt":"Residual neural networks are state-of-the-art deep learning models. Their continuous-depth analog, neural ordinary differential equations (ODEs), are also widely used. Despite their success, the link between the discrete and continuous models still lacks a solid mathematical foundation. In this article, we take a step in this direction by establishing an implicit regularization of deep residual networks towards neural ODEs, for nonlinear networks trained with gradient flow. We prove that if the network is initialized as a discretization of a neural ODE, then such a discretization holds through"},"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":"2309.01213","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-09-03T16:35:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"dd07a6312236063f231bf61256b511f92f54fbfaebd9fc77e4be4fc696c0b07d","abstract_canon_sha256":"b29925c012b17c6a91516314a2de18eafac9d25672afe840bf69170990faf831"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:40:08.670585Z","signature_b64":"4Bvd0P++Vx8/hEYDknpqn5d5Hp3kZJChvVLwS3MwQB5jRirSHhi0+kv36g8xlfe/bBtfAapwRLPv/OREpww8AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61ac8799ea4ebbace79c06cab396135c80ce60f39bce51231fdc9f4f4bad0f4b","last_reissued_at":"2026-07-05T08:40:08.670081Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:40:08.670081Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Implicit regularization of deep residual networks towards neural ODEs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"G\\'erard Biau, Michael E. Sander, Pierre Marion, Yu-Han Wu","submitted_at":"2023-09-03T16:35:59Z","abstract_excerpt":"Residual neural networks are state-of-the-art deep learning models. Their continuous-depth analog, neural ordinary differential equations (ODEs), are also widely used. Despite their success, the link between the discrete and continuous models still lacks a solid mathematical foundation. In this article, we take a step in this direction by establishing an implicit regularization of deep residual networks towards neural ODEs, for nonlinear networks trained with gradient flow. We prove that if the network is initialized as a discretization of a neural ODE, then such a discretization holds through"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.01213","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/2309.01213/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":"2309.01213","created_at":"2026-07-05T08:40:08.670139+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.01213v3","created_at":"2026-07-05T08:40:08.670139+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.01213","created_at":"2026-07-05T08:40:08.670139+00:00"},{"alias_kind":"pith_short_12","alias_value":"MGWIPGPKJ252","created_at":"2026-07-05T08:40:08.670139+00:00"},{"alias_kind":"pith_short_16","alias_value":"MGWIPGPKJ252ZZ44","created_at":"2026-07-05T08:40:08.670139+00:00"},{"alias_kind":"pith_short_8","alias_value":"MGWIPGPK","created_at":"2026-07-05T08:40:08.670139+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30930","citing_title":"SGD at the Edge of Stability: Stochastic Stabilization with Large Learning Rates","ref_index":184,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23778","citing_title":"The physics of AI weather models","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS","json":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS.json","graph_json":"https://pith.science/api/pith-number/MGWIPGPKJ252ZZ44A3FLHFQTLS/graph.json","events_json":"https://pith.science/api/pith-number/MGWIPGPKJ252ZZ44A3FLHFQTLS/events.json","paper":"https://pith.science/paper/MGWIPGPK"},"agent_actions":{"view_html":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS","download_json":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS.json","view_paper":"https://pith.science/paper/MGWIPGPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.01213&json=true","fetch_graph":"https://pith.science/api/pith-number/MGWIPGPKJ252ZZ44A3FLHFQTLS/graph.json","fetch_events":"https://pith.science/api/pith-number/MGWIPGPKJ252ZZ44A3FLHFQTLS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS/action/storage_attestation","attest_author":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS/action/author_attestation","sign_citation":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS/action/citation_signature","submit_replication":"https://pith.science/pith/MGWIPGPKJ252ZZ44A3FLHFQTLS/action/replication_record"}},"created_at":"2026-07-05T08:40:08.670139+00:00","updated_at":"2026-07-05T08:40:08.670139+00:00"}