{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:CKEBGIFR2AZQZFFRGM3247Y5BJ","short_pith_number":"pith:CKEBGIFR","schema_version":"1.0","canonical_sha256":"12881320b1d0330c94b13337ae7f1d0a6304d7d793791d988483c2e03de2aca0","source":{"kind":"arxiv","id":"1811.03962","version":5},"attestation_state":"computed","paper":{"title":"A Convergence Theory for Deep Learning via Over-Parameterization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","cs.NE","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Yuanzhi Li, Zeyuan Allen-Zhu, Zhao Song","submitted_at":"2018-11-09T15:16:13Z","abstract_excerpt":"Deep neural networks (DNNs) have demonstrated dominating performance in many fields; since AlexNet, networks used in practice are going wider and deeper. On the theoretical side, a long line of works has been focusing on training neural networks with one hidden layer. The theory of multi-layer networks remains largely unsettled.\n  In this work, we prove why stochastic gradient descent (SGD) can find $\\textit{global minima}$ on the training objective of DNNs in $\\textit{polynomial time}$. We only make two assumptions: the inputs are non-degenerate and the network is over-parameterized. The latt"},"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":"1811.03962","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-09T15:16:13Z","cross_cats_sorted":["cs.DS","cs.NE","math.OC","stat.ML"],"title_canon_sha256":"391002ecee5aac98737970e823a8165d44c3d535ef6a7be3c3ff412a44681a1e","abstract_canon_sha256":"2e3e26894853044740577b0dac69088bfdcb1bcaca753833d66233dca7f015e9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:17.268578Z","signature_b64":"TDiGnhkKkIyeVGIG/MgFGEX6mQgPHz44bS69iYxIv4hYviW4WNQ4OwHTHCQC5k7ABA04ZvaDjhaU2DP2U1vnDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"12881320b1d0330c94b13337ae7f1d0a6304d7d793791d988483c2e03de2aca0","last_reissued_at":"2026-05-17T23:43:17.267722Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:17.267722Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Convergence Theory for Deep Learning via Over-Parameterization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","cs.NE","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Yuanzhi Li, Zeyuan Allen-Zhu, Zhao Song","submitted_at":"2018-11-09T15:16:13Z","abstract_excerpt":"Deep neural networks (DNNs) have demonstrated dominating performance in many fields; since AlexNet, networks used in practice are going wider and deeper. On the theoretical side, a long line of works has been focusing on training neural networks with one hidden layer. The theory of multi-layer networks remains largely unsettled.\n  In this work, we prove why stochastic gradient descent (SGD) can find $\\textit{global minima}$ on the training objective of DNNs in $\\textit{polynomial time}$. We only make two assumptions: the inputs are non-degenerate and the network is over-parameterized. The latt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.03962","kind":"arxiv","version":5},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1811.03962","created_at":"2026-05-17T23:43:17.267863+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.03962v5","created_at":"2026-05-17T23:43:17.267863+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.03962","created_at":"2026-05-17T23:43:17.267863+00:00"},{"alias_kind":"pith_short_12","alias_value":"CKEBGIFR2AZQ","created_at":"2026-05-18T12:32:16.446611+00:00"},{"alias_kind":"pith_short_16","alias_value":"CKEBGIFR2AZQZFFR","created_at":"2026-05-18T12:32:16.446611+00:00"},{"alias_kind":"pith_short_8","alias_value":"CKEBGIFR","created_at":"2026-05-18T12:32:16.446611+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"1906.08654","citing_title":"ID3 Learns Juntas for Smoothed Product Distributions","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"1906.08899","citing_title":"Limitations of Lazy Training of Two-layers Neural Networks","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2102.11840","citing_title":"Convergence rates for gradient descent in the training of overparameterized artificial neural networks with piecewise affine activation","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2305.02657","citing_title":"On the Eigenvalue Decay Rates of a Class of Neural-Network Related Kernel Functions Defined on General Domains","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2509.20294","citing_title":"Alignment-Sensitive Minimax Rates for Spectral Algorithms with Learned Kernels","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09471","citing_title":"The Statistical Cost of Adaptation in Multi-Source Transfer Learning","ref_index":148,"is_internal_anchor":false},{"citing_arxiv_id":"2106.09685","citing_title":"LoRA: Low-Rank Adaptation of Large Language Models","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ","json":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ.json","graph_json":"https://pith.science/api/pith-number/CKEBGIFR2AZQZFFRGM3247Y5BJ/graph.json","events_json":"https://pith.science/api/pith-number/CKEBGIFR2AZQZFFRGM3247Y5BJ/events.json","paper":"https://pith.science/paper/CKEBGIFR"},"agent_actions":{"view_html":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ","download_json":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ.json","view_paper":"https://pith.science/paper/CKEBGIFR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.03962&json=true","fetch_graph":"https://pith.science/api/pith-number/CKEBGIFR2AZQZFFRGM3247Y5BJ/graph.json","fetch_events":"https://pith.science/api/pith-number/CKEBGIFR2AZQZFFRGM3247Y5BJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ/action/storage_attestation","attest_author":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ/action/author_attestation","sign_citation":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ/action/citation_signature","submit_replication":"https://pith.science/pith/CKEBGIFR2AZQZFFRGM3247Y5BJ/action/replication_record"}},"created_at":"2026-05-17T23:43:17.267863+00:00","updated_at":"2026-05-17T23:43:17.267863+00:00"}