{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:KDWDUY6S7I4IK3YRIIFJVXQF3J","short_pith_number":"pith:KDWDUY6S","schema_version":"1.0","canonical_sha256":"50ec3a63d2fa38856f11420a9ade05da50de24a9aa03fb9ee6dedcfd1ccd969c","source":{"kind":"arxiv","id":"1812.05994","version":1},"attestation_state":"computed","paper":{"title":"Products of Many Large Random Matrices and Gradients in Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math-ph","math.MP","stat.ML"],"primary_cat":"math.PR","authors_text":"Boris Hanin, Mihai Nica","submitted_at":"2018-12-14T15:59:34Z","abstract_excerpt":"We study products of random matrices in the regime where the number of terms and the size of the matrices simultaneously tend to infinity. Our main theorem is that the logarithm of the $\\ell_2$ norm of such a product applied to any fixed vector is asymptotically Gaussian. The fluctuations we find can be thought of as a finite temperature correction to the limit in which first the size and then the number of matrices tend to infinity. Depending on the scaling limit considered, the mean and variance of the limiting Gaussian depend only on either the first two or the first four moments of the mea"},"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":"1812.05994","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.PR","submitted_at":"2018-12-14T15:59:34Z","cross_cats_sorted":["math-ph","math.MP","stat.ML"],"title_canon_sha256":"51d091a9c01be40a8302e9604cb098dab505fe8a7a4f05cccc6d0c902fece9c9","abstract_canon_sha256":"c125974a3841f5d5938465745adf9c2dab0707a75f838408a41184ec8dd45b64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:36:17.609886Z","signature_b64":"ZTkvD9NBcF5Zk/+1BZp5n6Xz/nMj4Rv6jifizYnfa+r6cdpI0OWC3+m/vL31SONOjaAKtCXL0v982IjFJ96eBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50ec3a63d2fa38856f11420a9ade05da50de24a9aa03fb9ee6dedcfd1ccd969c","last_reissued_at":"2026-07-05T00:36:17.609408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:36:17.609408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Products of Many Large Random Matrices and Gradients in Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math-ph","math.MP","stat.ML"],"primary_cat":"math.PR","authors_text":"Boris Hanin, Mihai Nica","submitted_at":"2018-12-14T15:59:34Z","abstract_excerpt":"We study products of random matrices in the regime where the number of terms and the size of the matrices simultaneously tend to infinity. Our main theorem is that the logarithm of the $\\ell_2$ norm of such a product applied to any fixed vector is asymptotically Gaussian. The fluctuations we find can be thought of as a finite temperature correction to the limit in which first the size and then the number of matrices tend to infinity. Depending on the scaling limit considered, the mean and variance of the limiting Gaussian depend only on either the first two or the first four moments of the mea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.05994","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/1812.05994/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":"1812.05994","created_at":"2026-07-05T00:36:17.609464+00:00"},{"alias_kind":"arxiv_version","alias_value":"1812.05994v1","created_at":"2026-07-05T00:36:17.609464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.05994","created_at":"2026-07-05T00:36:17.609464+00:00"},{"alias_kind":"pith_short_12","alias_value":"KDWDUY6S7I4I","created_at":"2026-07-05T00:36:17.609464+00:00"},{"alias_kind":"pith_short_16","alias_value":"KDWDUY6S7I4IK3YR","created_at":"2026-07-05T00:36:17.609464+00:00"},{"alias_kind":"pith_short_8","alias_value":"KDWDUY6S","created_at":"2026-07-05T00:36:17.609464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02596","citing_title":"Spectral Asymptotics of Neural Network Loss Landscapes: An Exact Decomposition of the Curvature Exponent","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J","json":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J.json","graph_json":"https://pith.science/api/pith-number/KDWDUY6S7I4IK3YRIIFJVXQF3J/graph.json","events_json":"https://pith.science/api/pith-number/KDWDUY6S7I4IK3YRIIFJVXQF3J/events.json","paper":"https://pith.science/paper/KDWDUY6S"},"agent_actions":{"view_html":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J","download_json":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J.json","view_paper":"https://pith.science/paper/KDWDUY6S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1812.05994&json=true","fetch_graph":"https://pith.science/api/pith-number/KDWDUY6S7I4IK3YRIIFJVXQF3J/graph.json","fetch_events":"https://pith.science/api/pith-number/KDWDUY6S7I4IK3YRIIFJVXQF3J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J/action/storage_attestation","attest_author":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J/action/author_attestation","sign_citation":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J/action/citation_signature","submit_replication":"https://pith.science/pith/KDWDUY6S7I4IK3YRIIFJVXQF3J/action/replication_record"}},"created_at":"2026-07-05T00:36:17.609464+00:00","updated_at":"2026-07-05T00:36:17.609464+00:00"}