{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:G26AGQQSQYSL5RID3Z2SCDXM5G","short_pith_number":"pith:G26AGQQS","schema_version":"1.0","canonical_sha256":"36bc0342128624bec503de75210eece9bd39391782ada14b11c0be04121ea621","source":{"kind":"arxiv","id":"2211.11623","version":1},"attestation_state":"computed","paper":{"title":"Linear Stability Hypothesis and Rank Stratification for Nonlinear Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Leyang Zhang, Tao Luo, Yaoyu Zhang, Zhi-Qin John Xu, Zhiwei Bai, Zhongwang Zhang","submitted_at":"2022-11-21T16:27:25Z","abstract_excerpt":"Models with nonlinear architectures/parameterizations such as deep neural networks (DNNs) are well known for their mysteriously good generalization performance at overparameterization. In this work, we tackle this mystery from a novel perspective focusing on the transition of the target recovery/fitting accuracy as a function of the training data size. We propose a rank stratification for general nonlinear models to uncover a model rank as an \"effective size of parameters\" for each function in the function space of the corresponding model. Moreover, we establish a linear stability theory provi"},"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":"2211.11623","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2022-11-21T16:27:25Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"0e96f1393227fdd8ee59132f96c2d54a405e7c79b789f4a5b59dcdc2126645cc","abstract_canon_sha256":"db347ab4e38b167b8c067c6e239270fbc619c51d1dced0a94cf040566c004fca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:17:56.281851Z","signature_b64":"7Ao4x3ZWwA163jEptTlzVmSplnITPAgPEJ0BY/HiD9Gpa/no1If4oPcxZjif65uHRE18IkaG6Ge/Fo/tZRcOBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"36bc0342128624bec503de75210eece9bd39391782ada14b11c0be04121ea621","last_reissued_at":"2026-07-05T05:17:56.281345Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:17:56.281345Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Linear Stability Hypothesis and Rank Stratification for Nonlinear Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Leyang Zhang, Tao Luo, Yaoyu Zhang, Zhi-Qin John Xu, Zhiwei Bai, Zhongwang Zhang","submitted_at":"2022-11-21T16:27:25Z","abstract_excerpt":"Models with nonlinear architectures/parameterizations such as deep neural networks (DNNs) are well known for their mysteriously good generalization performance at overparameterization. In this work, we tackle this mystery from a novel perspective focusing on the transition of the target recovery/fitting accuracy as a function of the training data size. We propose a rank stratification for general nonlinear models to uncover a model rank as an \"effective size of parameters\" for each function in the function space of the corresponding model. Moreover, we establish a linear stability theory provi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.11623","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/2211.11623/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":"2211.11623","created_at":"2026-07-05T05:17:56.281411+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.11623v1","created_at":"2026-07-05T05:17:56.281411+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.11623","created_at":"2026-07-05T05:17:56.281411+00:00"},{"alias_kind":"pith_short_12","alias_value":"G26AGQQSQYSL","created_at":"2026-07-05T05:17:56.281411+00:00"},{"alias_kind":"pith_short_16","alias_value":"G26AGQQSQYSL5RID","created_at":"2026-07-05T05:17:56.281411+00:00"},{"alias_kind":"pith_short_8","alias_value":"G26AGQQS","created_at":"2026-07-05T05:17:56.281411+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07495","citing_title":"Second-Order Path Kernel Interpolation Formulas in Machine Learning","ref_index":83,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G","json":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G.json","graph_json":"https://pith.science/api/pith-number/G26AGQQSQYSL5RID3Z2SCDXM5G/graph.json","events_json":"https://pith.science/api/pith-number/G26AGQQSQYSL5RID3Z2SCDXM5G/events.json","paper":"https://pith.science/paper/G26AGQQS"},"agent_actions":{"view_html":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G","download_json":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G.json","view_paper":"https://pith.science/paper/G26AGQQS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.11623&json=true","fetch_graph":"https://pith.science/api/pith-number/G26AGQQSQYSL5RID3Z2SCDXM5G/graph.json","fetch_events":"https://pith.science/api/pith-number/G26AGQQSQYSL5RID3Z2SCDXM5G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G/action/storage_attestation","attest_author":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G/action/author_attestation","sign_citation":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G/action/citation_signature","submit_replication":"https://pith.science/pith/G26AGQQSQYSL5RID3Z2SCDXM5G/action/replication_record"}},"created_at":"2026-07-05T05:17:56.281411+00:00","updated_at":"2026-07-05T05:17:56.281411+00:00"}