{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BPR6OSIY2W5ITOJIWKZOSNA3YZ","short_pith_number":"pith:BPR6OSIY","schema_version":"1.0","canonical_sha256":"0be3e74918d5ba89b928b2b2e9341bc671b8f7ca39266edf42d7b36b7ef43f86","source":{"kind":"arxiv","id":"2312.11737","version":1},"attestation_state":"computed","paper":{"title":"Wide Deep Neural Networks with Gaussian Weights are Very Close to Gaussian Processes","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["math.PR","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Dario Trevisan","submitted_at":"2023-12-18T22:29:40Z","abstract_excerpt":"We establish novel rates for the Gaussian approximation of random deep neural networks with Gaussian parameters (weights and biases) and Lipschitz activation functions, in the wide limit. Our bounds apply for the joint output of a network evaluated any finite input set, provided a certain non-degeneracy condition of the infinite-width covariances holds. We demonstrate that the distance between the network output and the corresponding Gaussian approximation scales inversely with the width of the network, exhibiting faster convergence than the naive heuristic suggested by the central limit theor"},"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":"2312.11737","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"math.ST","submitted_at":"2023-12-18T22:29:40Z","cross_cats_sorted":["math.PR","stat.ML","stat.TH"],"title_canon_sha256":"a952c9486da6bf7ddef4f39e458da66286b8ef1a81ed69aacb7abb0bba09ba18","abstract_canon_sha256":"d68bed8e62a72a00f8abb9668b168291abd06a4558d69ae888de805c8792b0e3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:25:53.605049Z","signature_b64":"yJoZRl+07VskEqYCrgPhxMI+uaoR0Xh2NrR1hLewHOl3GU90oCHKdXMIYhxT3g9IpRAaBZlSEJtowRE2ldz8CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0be3e74918d5ba89b928b2b2e9341bc671b8f7ca39266edf42d7b36b7ef43f86","last_reissued_at":"2026-07-05T07:25:53.604655Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:25:53.604655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wide Deep Neural Networks with Gaussian Weights are Very Close to Gaussian Processes","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["math.PR","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Dario Trevisan","submitted_at":"2023-12-18T22:29:40Z","abstract_excerpt":"We establish novel rates for the Gaussian approximation of random deep neural networks with Gaussian parameters (weights and biases) and Lipschitz activation functions, in the wide limit. Our bounds apply for the joint output of a network evaluated any finite input set, provided a certain non-degeneracy condition of the infinite-width covariances holds. We demonstrate that the distance between the network output and the corresponding Gaussian approximation scales inversely with the width of the network, exhibiting faster convergence than the naive heuristic suggested by the central limit theor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.11737","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/2312.11737/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":"2312.11737","created_at":"2026-07-05T07:25:53.604713+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.11737v1","created_at":"2026-07-05T07:25:53.604713+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.11737","created_at":"2026-07-05T07:25:53.604713+00:00"},{"alias_kind":"pith_short_12","alias_value":"BPR6OSIY2W5I","created_at":"2026-07-05T07:25:53.604713+00:00"},{"alias_kind":"pith_short_16","alias_value":"BPR6OSIY2W5ITOJI","created_at":"2026-07-05T07:25:53.604713+00:00"},{"alias_kind":"pith_short_8","alias_value":"BPR6OSIY","created_at":"2026-07-05T07:25:53.604713+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":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30860","citing_title":"Bayesian Inference with Shaped Deep Non-linear MLPs","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2507.22095","citing_title":"Posterior Bayesian Neural Networks with Dependent Weights","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26898","citing_title":"Stochastic Scaling Limits and Synchronization by Noise in Deep Transformer Models","ref_index":51,"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":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ","json":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ.json","graph_json":"https://pith.science/api/pith-number/BPR6OSIY2W5ITOJIWKZOSNA3YZ/graph.json","events_json":"https://pith.science/api/pith-number/BPR6OSIY2W5ITOJIWKZOSNA3YZ/events.json","paper":"https://pith.science/paper/BPR6OSIY"},"agent_actions":{"view_html":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ","download_json":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ.json","view_paper":"https://pith.science/paper/BPR6OSIY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.11737&json=true","fetch_graph":"https://pith.science/api/pith-number/BPR6OSIY2W5ITOJIWKZOSNA3YZ/graph.json","fetch_events":"https://pith.science/api/pith-number/BPR6OSIY2W5ITOJIWKZOSNA3YZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ/action/storage_attestation","attest_author":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ/action/author_attestation","sign_citation":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ/action/citation_signature","submit_replication":"https://pith.science/pith/BPR6OSIY2W5ITOJIWKZOSNA3YZ/action/replication_record"}},"created_at":"2026-07-05T07:25:53.604713+00:00","updated_at":"2026-07-05T07:25:53.604713+00:00"}