{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:K5LBSRKRA5IO2VI433LYBCBD63","short_pith_number":"pith:K5LBSRKR","schema_version":"1.0","canonical_sha256":"57561945510750ed551cded7808823f6e6c165e2a306b0aa2d7be339b325e7a4","source":{"kind":"arxiv","id":"1904.09080","version":2},"attestation_state":"computed","paper":{"title":"Implicit regularization for deep neural networks driven by an Ornstein-Uhlenbeck like process","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Gregory Valiant, Guy Blanc, Neha Gupta, Paul Valiant","submitted_at":"2019-04-19T04:58:36Z","abstract_excerpt":"We consider networks, trained via stochastic gradient descent to minimize $\\ell_2$ loss, with the training labels perturbed by independent noise at each iteration. We characterize the behavior of the training dynamics near any parameter vector that achieves zero training error, in terms of an implicit regularization term corresponding to the sum over the data points, of the squared $\\ell_2$ norm of the gradient of the model with respect to the parameter vector, evaluated at each data point. This holds for networks of any connectivity, width, depth, and choice of activation function. We interpr"},"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":"1904.09080","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-04-19T04:58:36Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c6ccbec45e162da4943cf17720f67f71706acb70556be08fb2e10c595cc4acb7","abstract_canon_sha256":"ba82e04b499815a4b1a7350885d7062b8ba4f355b30e72836207ab02f7deffa3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:21:15.395807Z","signature_b64":"FWRZMEJV30k+sRedJtR1U0hwRQaR+wwo16oIDlZdci9kThDCDiMPqpAp27NfUnrBB9pCBwrsYsXDzbua3RgtDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"57561945510750ed551cded7808823f6e6c165e2a306b0aa2d7be339b325e7a4","last_reissued_at":"2026-07-05T01:21:15.395372Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:21:15.395372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Implicit regularization for deep neural networks driven by an Ornstein-Uhlenbeck like process","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Gregory Valiant, Guy Blanc, Neha Gupta, Paul Valiant","submitted_at":"2019-04-19T04:58:36Z","abstract_excerpt":"We consider networks, trained via stochastic gradient descent to minimize $\\ell_2$ loss, with the training labels perturbed by independent noise at each iteration. We characterize the behavior of the training dynamics near any parameter vector that achieves zero training error, in terms of an implicit regularization term corresponding to the sum over the data points, of the squared $\\ell_2$ norm of the gradient of the model with respect to the parameter vector, evaluated at each data point. This holds for networks of any connectivity, width, depth, and choice of activation function. We interpr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.09080","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/1904.09080/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":"1904.09080","created_at":"2026-07-05T01:21:15.395436+00:00"},{"alias_kind":"arxiv_version","alias_value":"1904.09080v2","created_at":"2026-07-05T01:21:15.395436+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.09080","created_at":"2026-07-05T01:21:15.395436+00:00"},{"alias_kind":"pith_short_12","alias_value":"K5LBSRKRA5IO","created_at":"2026-07-05T01:21:15.395436+00:00"},{"alias_kind":"pith_short_16","alias_value":"K5LBSRKRA5IO2VI4","created_at":"2026-07-05T01:21:15.395436+00:00"},{"alias_kind":"pith_short_8","alias_value":"K5LBSRKR","created_at":"2026-07-05T01:21:15.395436+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.01739","citing_title":"Grokking vs. Learning: Same Features, Different Encodings","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63","json":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63.json","graph_json":"https://pith.science/api/pith-number/K5LBSRKRA5IO2VI433LYBCBD63/graph.json","events_json":"https://pith.science/api/pith-number/K5LBSRKRA5IO2VI433LYBCBD63/events.json","paper":"https://pith.science/paper/K5LBSRKR"},"agent_actions":{"view_html":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63","download_json":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63.json","view_paper":"https://pith.science/paper/K5LBSRKR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1904.09080&json=true","fetch_graph":"https://pith.science/api/pith-number/K5LBSRKRA5IO2VI433LYBCBD63/graph.json","fetch_events":"https://pith.science/api/pith-number/K5LBSRKRA5IO2VI433LYBCBD63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63/action/storage_attestation","attest_author":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63/action/author_attestation","sign_citation":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63/action/citation_signature","submit_replication":"https://pith.science/pith/K5LBSRKRA5IO2VI433LYBCBD63/action/replication_record"}},"created_at":"2026-07-05T01:21:15.395436+00:00","updated_at":"2026-07-05T01:21:15.395436+00:00"}