{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:52NKXIGFARBDITXTQAT57QLXEH","short_pith_number":"pith:52NKXIGF","schema_version":"1.0","canonical_sha256":"ee9aaba0c50442344ef38027dfc17721d4f499f2d7469e681fbe6bae68a05969","source":{"kind":"arxiv","id":"2311.14609","version":1},"attestation_state":"computed","paper":{"title":"Analysis of the expected $L_2$ error of an over-parametrized deep neural network estimate learned by gradient descent without regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Michael Kohler, Selina Drews","submitted_at":"2023-11-24T17:04:21Z","abstract_excerpt":"Recent results show that estimates defined by over-parametrized deep neural networks learned by applying gradient descent to a regularized empirical $L_2$ risk are universally consistent and achieve good rates of convergence. In this paper, we show that the regularization term is not necessary to obtain similar results. In the case of a suitably chosen initialization of the network, a suitable number of gradient descent steps, and a suitable step size we show that an estimate without a regularization term is universally consistent for bounded predictor variables. Additionally, we show that if "},"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":"2311.14609","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2023-11-24T17:04:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ada92a62f56834a5f46e46695b0479626cf1767071a1c410c168e6b06bffa872","abstract_canon_sha256":"607f25cbd986d4f7421a806aab7250a0e9bb5fab5a3d1c0260e78db1a2e499d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:16:21.765454Z","signature_b64":"2XXD5q9WEWbG/FKvz0Ost4uCGzSEqMr0q98XDVEq2IikI1KxKW06HYrGHNbYzuYBGt6RWYN7/BtuTKEVSVrWDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ee9aaba0c50442344ef38027dfc17721d4f499f2d7469e681fbe6bae68a05969","last_reissued_at":"2026-07-05T07:16:21.764923Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:16:21.764923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Analysis of the expected $L_2$ error of an over-parametrized deep neural network estimate learned by gradient descent without regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Michael Kohler, Selina Drews","submitted_at":"2023-11-24T17:04:21Z","abstract_excerpt":"Recent results show that estimates defined by over-parametrized deep neural networks learned by applying gradient descent to a regularized empirical $L_2$ risk are universally consistent and achieve good rates of convergence. In this paper, we show that the regularization term is not necessary to obtain similar results. In the case of a suitably chosen initialization of the network, a suitable number of gradient descent steps, and a suitable step size we show that an estimate without a regularization term is universally consistent for bounded predictor variables. Additionally, we show that if "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.14609","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/2311.14609/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":"2311.14609","created_at":"2026-07-05T07:16:21.764995+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.14609v1","created_at":"2026-07-05T07:16:21.764995+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.14609","created_at":"2026-07-05T07:16:21.764995+00:00"},{"alias_kind":"pith_short_12","alias_value":"52NKXIGFARBD","created_at":"2026-07-05T07:16:21.764995+00:00"},{"alias_kind":"pith_short_16","alias_value":"52NKXIGFARBDITXT","created_at":"2026-07-05T07:16:21.764995+00:00"},{"alias_kind":"pith_short_8","alias_value":"52NKXIGF","created_at":"2026-07-05T07:16:21.764995+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06772","citing_title":"Minimax-Optimal Generalization Bounds for Smooth Deep Neural Networks Trained by (Stochastic) Gradient Descent","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH","json":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH.json","graph_json":"https://pith.science/api/pith-number/52NKXIGFARBDITXTQAT57QLXEH/graph.json","events_json":"https://pith.science/api/pith-number/52NKXIGFARBDITXTQAT57QLXEH/events.json","paper":"https://pith.science/paper/52NKXIGF"},"agent_actions":{"view_html":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH","download_json":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH.json","view_paper":"https://pith.science/paper/52NKXIGF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.14609&json=true","fetch_graph":"https://pith.science/api/pith-number/52NKXIGFARBDITXTQAT57QLXEH/graph.json","fetch_events":"https://pith.science/api/pith-number/52NKXIGFARBDITXTQAT57QLXEH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH/action/storage_attestation","attest_author":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH/action/author_attestation","sign_citation":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH/action/citation_signature","submit_replication":"https://pith.science/pith/52NKXIGFARBDITXTQAT57QLXEH/action/replication_record"}},"created_at":"2026-07-05T07:16:21.764995+00:00","updated_at":"2026-07-05T07:16:21.764995+00:00"}