{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SECUPJS2YV46JWG4YU75NZLW2H","short_pith_number":"pith:SECUPJS2","canonical_record":{"source":{"id":"2405.13456","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-05-22T08:58:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f043570a11ff8bf586756ae84eb3391df3dac855242bf5bbd3d71a3f86c9a55d","abstract_canon_sha256":"64cca06a74eb424bacd9cdfffb3bbaff6eb0323d79c56b60b6c2823a725b3363"},"schema_version":"1.0"},"canonical_sha256":"910547a65ac579e4d8dcc53fd6e576d1f3eae8f2c695178b8820cc2f61aba7e0","source":{"kind":"arxiv","id":"2405.13456","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.13456","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"arxiv_version","alias_value":"2405.13456v2","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13456","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"pith_short_12","alias_value":"SECUPJS2YV46","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"pith_short_16","alias_value":"SECUPJS2YV46JWG4","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"pith_short_8","alias_value":"SECUPJS2","created_at":"2026-07-05T09:27:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SECUPJS2YV46JWG4YU75NZLW2H","target":"record","payload":{"canonical_record":{"source":{"id":"2405.13456","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-05-22T08:58:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f043570a11ff8bf586756ae84eb3391df3dac855242bf5bbd3d71a3f86c9a55d","abstract_canon_sha256":"64cca06a74eb424bacd9cdfffb3bbaff6eb0323d79c56b60b6c2823a725b3363"},"schema_version":"1.0"},"canonical_sha256":"910547a65ac579e4d8dcc53fd6e576d1f3eae8f2c695178b8820cc2f61aba7e0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:00.636458Z","signature_b64":"8yLzCS3m2Y7i7USR2o+u851a93zNHMHsLeZcFbkJXIdW6rWgmpqtzKo8i1JsSBk/0GCsH5IRqgsWkxRkPh1iCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"910547a65ac579e4d8dcc53fd6e576d1f3eae8f2c695178b8820cc2f61aba7e0","last_reissued_at":"2026-07-05T09:27:00.635965Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:00.635965Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.13456","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:27:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"swvg5wI0X0Te1R5AU//wfeTG/lbSnjPAoQeOcUqfsGR+VBxNsxspdM+IDcq1/DkEM4GgPXD6rQ9rUBQE+JB7CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T17:55:47.647731Z"},"content_sha256":"3428f3dbc38863eb7eab1cd3ca7dabfebfd3844fb5390f61670d30e02762d0ca","schema_version":"1.0","event_id":"sha256:3428f3dbc38863eb7eab1cd3ca7dabfebfd3844fb5390f61670d30e02762d0ca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SECUPJS2YV46JWG4YU75NZLW2H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Deep linear networks for regression are implicitly regularized towards flat minima","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"L\\'ena\\\"ic Chizat, Pierre Marion","submitted_at":"2024-05-22T08:58:51Z","abstract_excerpt":"The largest eigenvalue of the Hessian, or sharpness, of neural networks is a key quantity to understand their optimization dynamics. In this paper, we study the sharpness of deep linear networks for univariate regression. Minimizers can have arbitrarily large sharpness, but not an arbitrarily small one. Indeed, we show a lower bound on the sharpness of minimizers, which grows linearly with depth. We then study the properties of the minimizer found by gradient flow, which is the limit of gradient descent with vanishing learning rate. We show an implicit regularization towards flat minima: the s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13456","kind":"arxiv","version":2},"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/2405.13456/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:27:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qLPxIjXSIVASDqEu49PcC3yn1OMWufWH9zd6zDuFEYi00Crb85Qprs/gIoWRqqL5r8lnIJwZp/720rGZlAlDCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T17:55:47.648273Z"},"content_sha256":"2daca9019cdd661886ff4966e5ce16b62a51883b694b0403be0346f65783cdc1","schema_version":"1.0","event_id":"sha256:2daca9019cdd661886ff4966e5ce16b62a51883b694b0403be0346f65783cdc1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SECUPJS2YV46JWG4YU75NZLW2H/bundle.json","state_url":"https://pith.science/pith/SECUPJS2YV46JWG4YU75NZLW2H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SECUPJS2YV46JWG4YU75NZLW2H/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T17:55:47Z","links":{"resolver":"https://pith.science/pith/SECUPJS2YV46JWG4YU75NZLW2H","bundle":"https://pith.science/pith/SECUPJS2YV46JWG4YU75NZLW2H/bundle.json","state":"https://pith.science/pith/SECUPJS2YV46JWG4YU75NZLW2H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SECUPJS2YV46JWG4YU75NZLW2H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SECUPJS2YV46JWG4YU75NZLW2H","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"64cca06a74eb424bacd9cdfffb3bbaff6eb0323d79c56b60b6c2823a725b3363","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-05-22T08:58:51Z","title_canon_sha256":"f043570a11ff8bf586756ae84eb3391df3dac855242bf5bbd3d71a3f86c9a55d"},"schema_version":"1.0","source":{"id":"2405.13456","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.13456","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"arxiv_version","alias_value":"2405.13456v2","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13456","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"pith_short_12","alias_value":"SECUPJS2YV46","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"pith_short_16","alias_value":"SECUPJS2YV46JWG4","created_at":"2026-07-05T09:27:00Z"},{"alias_kind":"pith_short_8","alias_value":"SECUPJS2","created_at":"2026-07-05T09:27:00Z"}],"graph_snapshots":[{"event_id":"sha256:2daca9019cdd661886ff4966e5ce16b62a51883b694b0403be0346f65783cdc1","target":"graph","created_at":"2026-07-05T09:27:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2405.13456/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The largest eigenvalue of the Hessian, or sharpness, of neural networks is a key quantity to understand their optimization dynamics. In this paper, we study the sharpness of deep linear networks for univariate regression. Minimizers can have arbitrarily large sharpness, but not an arbitrarily small one. Indeed, we show a lower bound on the sharpness of minimizers, which grows linearly with depth. We then study the properties of the minimizer found by gradient flow, which is the limit of gradient descent with vanishing learning rate. We show an implicit regularization towards flat minima: the s","authors_text":"L\\'ena\\\"ic Chizat, Pierre Marion","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-05-22T08:58:51Z","title":"Deep linear networks for regression are implicitly regularized towards flat minima"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13456","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3428f3dbc38863eb7eab1cd3ca7dabfebfd3844fb5390f61670d30e02762d0ca","target":"record","created_at":"2026-07-05T09:27:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"64cca06a74eb424bacd9cdfffb3bbaff6eb0323d79c56b60b6c2823a725b3363","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-05-22T08:58:51Z","title_canon_sha256":"f043570a11ff8bf586756ae84eb3391df3dac855242bf5bbd3d71a3f86c9a55d"},"schema_version":"1.0","source":{"id":"2405.13456","kind":"arxiv","version":2}},"canonical_sha256":"910547a65ac579e4d8dcc53fd6e576d1f3eae8f2c695178b8820cc2f61aba7e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"910547a65ac579e4d8dcc53fd6e576d1f3eae8f2c695178b8820cc2f61aba7e0","first_computed_at":"2026-07-05T09:27:00.635965Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:27:00.635965Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8yLzCS3m2Y7i7USR2o+u851a93zNHMHsLeZcFbkJXIdW6rWgmpqtzKo8i1JsSBk/0GCsH5IRqgsWkxRkPh1iCw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:27:00.636458Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.13456","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3428f3dbc38863eb7eab1cd3ca7dabfebfd3844fb5390f61670d30e02762d0ca","sha256:2daca9019cdd661886ff4966e5ce16b62a51883b694b0403be0346f65783cdc1"],"state_sha256":"14e7b6b41f8dd33ec1cb1360a2ddfb5a1febe701dabb754bf6b6a97eb9a45ee1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yV2vnQYcKYWLevk2ukqlgFGYlzXvJOv9UKHdFeFvAHUsElGlyCLTSMt6ulQvgw48ZVxjwZzGnpNZ63bAj9O+Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T17:55:47.652458Z","bundle_sha256":"671c87f89cda854587dedc818fc0b13062fa2f9771251960ebcea51729f74ff4"}}