{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JGHAAEGDH7QQKIKNWL27UUD37Q","short_pith_number":"pith:JGHAAEGD","schema_version":"1.0","canonical_sha256":"498e0010c33fe105214db2f5fa507bfc0588c1731ebeaf3016cd93051e9d8238","source":{"kind":"arxiv","id":"2412.15911","version":1},"attestation_state":"computed","paper":{"title":"Kernel shape renormalization explains output-output correlations in finite Bayesian one-hidden-layer networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cond-mat.dis-nn","authors_text":"A. Vezzani, L. Giambagli, P. Baglioni, P. Rotondo, R. Burioni, R. Pacelli","submitted_at":"2024-12-20T14:01:16Z","abstract_excerpt":"Finite-width one hidden layer networks with multiple neurons in the readout layer display non-trivial output-output correlations that vanish in the lazy-training infinite-width limit. In this manuscript we leverage recent progress in the proportional limit of Bayesian deep learning (that is the limit where the size of the training set $P$ and the width of the hidden layers $N$ are taken to infinity keeping their ratio $\\alpha = P/N$ finite) to rationalize this empirical evidence. In particular, we show that output-output correlations in finite fully-connected networks are taken into account by"},"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":"2412.15911","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cond-mat.dis-nn","submitted_at":"2024-12-20T14:01:16Z","cross_cats_sorted":[],"title_canon_sha256":"3235caffa7a2a0aeb7059425558bcb5ad1b01f9421f9cbad6f47168db32177d9","abstract_canon_sha256":"d0f44a5419938b92752ce31b6500c4ba5f649203dbc999caf0418727e7337dd2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:31.967759Z","signature_b64":"UnLE8DSw2hnfTfu2b2+v023exyyZYXPsyJ5isCO/Ch9ekt0AwVzsKYlpYun4cTIB7jBtJqrST6o3rSxWIV4kAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"498e0010c33fe105214db2f5fa507bfc0588c1731ebeaf3016cd93051e9d8238","last_reissued_at":"2026-07-05T09:52:31.967259Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:31.967259Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Kernel shape renormalization explains output-output correlations in finite Bayesian one-hidden-layer networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cond-mat.dis-nn","authors_text":"A. Vezzani, L. Giambagli, P. Baglioni, P. Rotondo, R. Burioni, R. Pacelli","submitted_at":"2024-12-20T14:01:16Z","abstract_excerpt":"Finite-width one hidden layer networks with multiple neurons in the readout layer display non-trivial output-output correlations that vanish in the lazy-training infinite-width limit. In this manuscript we leverage recent progress in the proportional limit of Bayesian deep learning (that is the limit where the size of the training set $P$ and the width of the hidden layers $N$ are taken to infinity keeping their ratio $\\alpha = P/N$ finite) to rationalize this empirical evidence. In particular, we show that output-output correlations in finite fully-connected networks are taken into account by"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15911","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/2412.15911/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":"2412.15911","created_at":"2026-07-05T09:52:31.967321+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.15911v1","created_at":"2026-07-05T09:52:31.967321+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15911","created_at":"2026-07-05T09:52:31.967321+00:00"},{"alias_kind":"pith_short_12","alias_value":"JGHAAEGDH7QQ","created_at":"2026-07-05T09:52:31.967321+00:00"},{"alias_kind":"pith_short_16","alias_value":"JGHAAEGDH7QQKIKN","created_at":"2026-07-05T09:52:31.967321+00:00"},{"alias_kind":"pith_short_8","alias_value":"JGHAAEGD","created_at":"2026-07-05T09:52:31.967321+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/JGHAAEGDH7QQKIKNWL27UUD37Q","json":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q.json","graph_json":"https://pith.science/api/pith-number/JGHAAEGDH7QQKIKNWL27UUD37Q/graph.json","events_json":"https://pith.science/api/pith-number/JGHAAEGDH7QQKIKNWL27UUD37Q/events.json","paper":"https://pith.science/paper/JGHAAEGD"},"agent_actions":{"view_html":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q","download_json":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q.json","view_paper":"https://pith.science/paper/JGHAAEGD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.15911&json=true","fetch_graph":"https://pith.science/api/pith-number/JGHAAEGDH7QQKIKNWL27UUD37Q/graph.json","fetch_events":"https://pith.science/api/pith-number/JGHAAEGDH7QQKIKNWL27UUD37Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q/action/storage_attestation","attest_author":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q/action/author_attestation","sign_citation":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q/action/citation_signature","submit_replication":"https://pith.science/pith/JGHAAEGDH7QQKIKNWL27UUD37Q/action/replication_record"}},"created_at":"2026-07-05T09:52:31.967321+00:00","updated_at":"2026-07-05T09:52:31.967321+00:00"}