{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:P57Y6BR7YR4XWRMGFI7XTHN2RI","short_pith_number":"pith:P57Y6BR7","schema_version":"1.0","canonical_sha256":"7f7f8f063fc4797b45862a3f799dba8a0655cc6f9d4de99c1a4c985b19beb551","source":{"kind":"arxiv","id":"2203.12967","version":1},"attestation_state":"computed","paper":{"title":"Extended critical regimes of deep neural networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cond-mat.stat-mech","cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Asem Wardak, Cheng Kevin Qu, Pulin Gong","submitted_at":"2022-03-24T10:15:50Z","abstract_excerpt":"Deep neural networks (DNNs) have been successfully applied to many real-world problems, but a complete understanding of their dynamical and computational principles is still lacking. Conventional theoretical frameworks for analysing DNNs often assume random networks with coupling weights obeying Gaussian statistics. However, non-Gaussian, heavy-tailed coupling is a ubiquitous phenomenon in DNNs. Here, by weaving together theories of heavy-tailed random matrices and non-equilibrium statistical physics, we develop a new type of mean field theory for DNNs which predicts that heavy-tailed weights "},"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":"2203.12967","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-24T10:15:50Z","cross_cats_sorted":["cond-mat.dis-nn","cond-mat.stat-mech","cs.AI","stat.ML"],"title_canon_sha256":"96511515413db8124645b01f20751a4fcecabc8d25bbbd65c8c7232d56769cb4","abstract_canon_sha256":"062fcca69409f910fb748948297012e61bd3ce6d399465a58355c3e6f9f2e11c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:08:19.378470Z","signature_b64":"q2rHJkpRfwAqehgD6JI72Ryk07MVVahO/AHH4kqY+hom64Vby7LPSnSkXdjukBS08s8jwQDjNDmHq9hyPec3Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f7f8f063fc4797b45862a3f799dba8a0655cc6f9d4de99c1a4c985b19beb551","last_reissued_at":"2026-07-05T04:08:19.378054Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:08:19.378054Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Extended critical regimes of deep neural networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cond-mat.stat-mech","cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Asem Wardak, Cheng Kevin Qu, Pulin Gong","submitted_at":"2022-03-24T10:15:50Z","abstract_excerpt":"Deep neural networks (DNNs) have been successfully applied to many real-world problems, but a complete understanding of their dynamical and computational principles is still lacking. Conventional theoretical frameworks for analysing DNNs often assume random networks with coupling weights obeying Gaussian statistics. However, non-Gaussian, heavy-tailed coupling is a ubiquitous phenomenon in DNNs. Here, by weaving together theories of heavy-tailed random matrices and non-equilibrium statistical physics, we develop a new type of mean field theory for DNNs which predicts that heavy-tailed weights "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.12967","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/2203.12967/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":"2203.12967","created_at":"2026-07-05T04:08:19.378108+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.12967v1","created_at":"2026-07-05T04:08:19.378108+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.12967","created_at":"2026-07-05T04:08:19.378108+00:00"},{"alias_kind":"pith_short_12","alias_value":"P57Y6BR7YR4X","created_at":"2026-07-05T04:08:19.378108+00:00"},{"alias_kind":"pith_short_16","alias_value":"P57Y6BR7YR4XWRMG","created_at":"2026-07-05T04:08:19.378108+00:00"},{"alias_kind":"pith_short_8","alias_value":"P57Y6BR7","created_at":"2026-07-05T04:08:19.378108+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14258","citing_title":"Dynamics of the Transformer Residual Stream: Coupling Spectral Geometry to Network Topology","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI","json":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI.json","graph_json":"https://pith.science/api/pith-number/P57Y6BR7YR4XWRMGFI7XTHN2RI/graph.json","events_json":"https://pith.science/api/pith-number/P57Y6BR7YR4XWRMGFI7XTHN2RI/events.json","paper":"https://pith.science/paper/P57Y6BR7"},"agent_actions":{"view_html":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI","download_json":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI.json","view_paper":"https://pith.science/paper/P57Y6BR7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.12967&json=true","fetch_graph":"https://pith.science/api/pith-number/P57Y6BR7YR4XWRMGFI7XTHN2RI/graph.json","fetch_events":"https://pith.science/api/pith-number/P57Y6BR7YR4XWRMGFI7XTHN2RI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI/action/storage_attestation","attest_author":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI/action/author_attestation","sign_citation":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI/action/citation_signature","submit_replication":"https://pith.science/pith/P57Y6BR7YR4XWRMGFI7XTHN2RI/action/replication_record"}},"created_at":"2026-07-05T04:08:19.378108+00:00","updated_at":"2026-07-05T04:08:19.378108+00:00"}