{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EOXVRSM6AAKCVQMDFIRYOOX4YD","short_pith_number":"pith:EOXVRSM6","schema_version":"1.0","canonical_sha256":"23af58c99e00142ac1832a23873afcc0d7d8e202706f7e99ca1d72ea30b36077","source":{"kind":"arxiv","id":"2504.08335","version":2},"attestation_state":"computed","paper":{"title":"Entropic bounds for conditionally Gaussian vectors and applications to neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"math.PR","authors_text":"Giovanni Peccati, Lucia Celli","submitted_at":"2025-04-11T08:00:37Z","abstract_excerpt":"Using entropic inequalities from information theory, we provide new bounds on the total variation and 2-Wasserstein distances between a conditionally Gaussian law and a Gaussian law with invertible covariance matrix. We apply our results to quantify the speed of convergence to Gaussian of a randomly initialized fully connected neural network and its derivatives - evaluated in a finite number of inputs - when the initialization is Gaussian and the sizes of the inner layers diverge to infinity. Our results require mild assumptions on the activation function, and allow one to recover optimal rate"},"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":"2504.08335","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.PR","submitted_at":"2025-04-11T08:00:37Z","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"title_canon_sha256":"027d3974f8accc07a3e37afa4bffd7b891c81963c76621e88a746fc32495ccdd","abstract_canon_sha256":"4f20415ed1fda4885c7eaf1d58c2ca90e80bdea18160cc3e999e945b1af75a11"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:58.084873Z","signature_b64":"z1vE5j/Hj553c6wc5EIWXTYKKDvWv8eUONS5hFgyUMWoB+LilXlpw7O1XvkvVAGWbzscBgCG0+azPBEqKfniCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23af58c99e00142ac1832a23873afcc0d7d8e202706f7e99ca1d72ea30b36077","last_reissued_at":"2026-07-05T11:14:58.084206Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:58.084206Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Entropic bounds for conditionally Gaussian vectors and applications to neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"math.PR","authors_text":"Giovanni Peccati, Lucia Celli","submitted_at":"2025-04-11T08:00:37Z","abstract_excerpt":"Using entropic inequalities from information theory, we provide new bounds on the total variation and 2-Wasserstein distances between a conditionally Gaussian law and a Gaussian law with invertible covariance matrix. We apply our results to quantify the speed of convergence to Gaussian of a randomly initialized fully connected neural network and its derivatives - evaluated in a finite number of inputs - when the initialization is Gaussian and the sizes of the inner layers diverge to infinity. Our results require mild assumptions on the activation function, and allow one to recover optimal rate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.08335","kind":"arxiv","version":2},"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/2504.08335/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":"2504.08335","created_at":"2026-07-05T11:14:58.084264+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.08335v2","created_at":"2026-07-05T11:14:58.084264+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.08335","created_at":"2026-07-05T11:14:58.084264+00:00"},{"alias_kind":"pith_short_12","alias_value":"EOXVRSM6AAKC","created_at":"2026-07-05T11:14:58.084264+00:00"},{"alias_kind":"pith_short_16","alias_value":"EOXVRSM6AAKCVQMD","created_at":"2026-07-05T11:14:58.084264+00:00"},{"alias_kind":"pith_short_8","alias_value":"EOXVRSM6","created_at":"2026-07-05T11:14:58.084264+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30831","citing_title":"Geometric Dyson Brownian Motions and the Free Log-Normal Limit for a Non-Square Gaussian Matrix Product","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24072","citing_title":"Optimal Non-Asymptotic Edgeworth Expansions for Multivariate Neural Network Outputs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2507.22095","citing_title":"Posterior Bayesian Neural Networks with Dependent Weights","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19738","citing_title":"Phase Transitions in the Fluctuations of Functionals of Random Neural Networks","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02771","citing_title":"Universality in Deep Neural Networks: An approach via the Lindeberg exchange principle","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD","json":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD.json","graph_json":"https://pith.science/api/pith-number/EOXVRSM6AAKCVQMDFIRYOOX4YD/graph.json","events_json":"https://pith.science/api/pith-number/EOXVRSM6AAKCVQMDFIRYOOX4YD/events.json","paper":"https://pith.science/paper/EOXVRSM6"},"agent_actions":{"view_html":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD","download_json":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD.json","view_paper":"https://pith.science/paper/EOXVRSM6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.08335&json=true","fetch_graph":"https://pith.science/api/pith-number/EOXVRSM6AAKCVQMDFIRYOOX4YD/graph.json","fetch_events":"https://pith.science/api/pith-number/EOXVRSM6AAKCVQMDFIRYOOX4YD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD/action/storage_attestation","attest_author":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD/action/author_attestation","sign_citation":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD/action/citation_signature","submit_replication":"https://pith.science/pith/EOXVRSM6AAKCVQMDFIRYOOX4YD/action/replication_record"}},"created_at":"2026-07-05T11:14:58.084264+00:00","updated_at":"2026-07-05T11:14:58.084264+00:00"}