{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VEHMXFAHMK5DCY6RGMIDNI4S4O","short_pith_number":"pith:VEHMXFAH","schema_version":"1.0","canonical_sha256":"a90ecb940762ba3163d1331036a392e382e5ec5bf8919ccc07536b4623362826","source":{"kind":"arxiv","id":"2605.27991","version":1},"attestation_state":"computed","paper":{"title":"Deep Neural Network Training as Random Effects: An Optimization-Inference Duality","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Lin Liu, Minhao Yao, Ruoyu Wang, Xihong Lin, Zhonghua Liu","submitted_at":"2026-05-27T05:32:24Z","abstract_excerpt":"Deep neural networks (DNNs) have achieved remarkable empirical success, yet their training dynamics remain understood mainly from optimization rather than statistical principles. Here we develop a statistical framework for DNN training in the over-parameterized regime by showing that the prediction induced by continuous-time neural tangent kernel (NTK) gradient flow is exactly equivalent to that from a classical random-effects model. In this framework, training time acts as a variance component, or equivalently an empirical Bayes covariance hyperparameter, governing the allocation of variation"},"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":"2605.27991","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-27T05:32:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0855c97187bbd586933820a37dac95bf976751eec5e222d18a82ecd528e8e96d","abstract_canon_sha256":"a36b2a08c2bb3df86e52c61a8b89f1a460b59da57857cd5ed56c630e9d0f2265"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:55.423190Z","signature_b64":"135ClbOpFNQv42oNezuR24GcWx4XMTQxz0iTlhQKgGRy71jlU4jkJUkbDhcinfIjZJbdok9CQWzkNLHQSibxDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a90ecb940762ba3163d1331036a392e382e5ec5bf8919ccc07536b4623362826","last_reissued_at":"2026-05-28T01:04:55.422728Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:55.422728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Neural Network Training as Random Effects: An Optimization-Inference Duality","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Lin Liu, Minhao Yao, Ruoyu Wang, Xihong Lin, Zhonghua Liu","submitted_at":"2026-05-27T05:32:24Z","abstract_excerpt":"Deep neural networks (DNNs) have achieved remarkable empirical success, yet their training dynamics remain understood mainly from optimization rather than statistical principles. Here we develop a statistical framework for DNN training in the over-parameterized regime by showing that the prediction induced by continuous-time neural tangent kernel (NTK) gradient flow is exactly equivalent to that from a classical random-effects model. In this framework, training time acts as a variance component, or equivalently an empirical Bayes covariance hyperparameter, governing the allocation of variation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27991","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/2605.27991/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":"2605.27991","created_at":"2026-05-28T01:04:55.422796+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27991v1","created_at":"2026-05-28T01:04:55.422796+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27991","created_at":"2026-05-28T01:04:55.422796+00:00"},{"alias_kind":"pith_short_12","alias_value":"VEHMXFAHMK5D","created_at":"2026-05-28T01:04:55.422796+00:00"},{"alias_kind":"pith_short_16","alias_value":"VEHMXFAHMK5DCY6R","created_at":"2026-05-28T01:04:55.422796+00:00"},{"alias_kind":"pith_short_8","alias_value":"VEHMXFAH","created_at":"2026-05-28T01:04:55.422796+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O","json":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O.json","graph_json":"https://pith.science/api/pith-number/VEHMXFAHMK5DCY6RGMIDNI4S4O/graph.json","events_json":"https://pith.science/api/pith-number/VEHMXFAHMK5DCY6RGMIDNI4S4O/events.json","paper":"https://pith.science/paper/VEHMXFAH"},"agent_actions":{"view_html":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O","download_json":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O.json","view_paper":"https://pith.science/paper/VEHMXFAH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27991&json=true","fetch_graph":"https://pith.science/api/pith-number/VEHMXFAHMK5DCY6RGMIDNI4S4O/graph.json","fetch_events":"https://pith.science/api/pith-number/VEHMXFAHMK5DCY6RGMIDNI4S4O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O/action/storage_attestation","attest_author":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O/action/author_attestation","sign_citation":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O/action/citation_signature","submit_replication":"https://pith.science/pith/VEHMXFAHMK5DCY6RGMIDNI4S4O/action/replication_record"}},"created_at":"2026-05-28T01:04:55.422796+00:00","updated_at":"2026-05-28T01:04:55.422796+00:00"}