{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:XH42SYFBYF3MDDJMFO6QFRCU7I","short_pith_number":"pith:XH42SYFB","schema_version":"1.0","canonical_sha256":"b9f9a960a1c176c18d2c2bbd02c454fa124282a731d2acc2ba91b822e2a49f02","source":{"kind":"arxiv","id":"2204.10425","version":1},"attestation_state":"computed","paper":{"title":"Spectrum of inner-product kernel matrices in the polynomial regime and multiple descent phenomenon in kernel ridge regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Theodor Misiakiewicz","submitted_at":"2022-04-21T22:20:52Z","abstract_excerpt":"We study the spectrum of inner-product kernel matrices, i.e., $n \\times n$ matrices with entries $h (\\langle \\textbf{x}_i ,\\textbf{x}_j \\rangle/d)$ where the $( \\textbf{x}_i)_{i \\leq n}$ are i.i.d.~random covariates in $\\mathbb{R}^d$. In the linear high-dimensional regime $n \\asymp d$, it was shown that these matrices are well approximated by their linearization, which simplifies into the sum of a rescaled Wishart matrix and identity matrix. In this paper, we generalize this decomposition to the polynomial high-dimensional regime $n \\asymp d^\\ell,\\ell \\in \\mathbb{N}$, for data uniformly distri"},"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":"2204.10425","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2022-04-21T22:20:52Z","cross_cats_sorted":["stat.ML","stat.TH"],"title_canon_sha256":"9ccedeb5795b6a2d97e401ce49350d8b712b6aa0643d6e470f72e41b394282e2","abstract_canon_sha256":"6c7e637335212da56cad8982826710f325c243e6f04ebbbf4e19e06b2b75f9b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:16:56.731773Z","signature_b64":"lNeeRRTm33rkRoKigMuXpQSp/bhwrA0nq5ldaEvVGz0yWit0oM/cgFuiMAGG/NsWnnnAwxK35XXghDOKP2q0CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9f9a960a1c176c18d2c2bbd02c454fa124282a731d2acc2ba91b822e2a49f02","last_reissued_at":"2026-07-05T04:16:56.731379Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:16:56.731379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spectrum of inner-product kernel matrices in the polynomial regime and multiple descent phenomenon in kernel ridge regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Theodor Misiakiewicz","submitted_at":"2022-04-21T22:20:52Z","abstract_excerpt":"We study the spectrum of inner-product kernel matrices, i.e., $n \\times n$ matrices with entries $h (\\langle \\textbf{x}_i ,\\textbf{x}_j \\rangle/d)$ where the $( \\textbf{x}_i)_{i \\leq n}$ are i.i.d.~random covariates in $\\mathbb{R}^d$. In the linear high-dimensional regime $n \\asymp d$, it was shown that these matrices are well approximated by their linearization, which simplifies into the sum of a rescaled Wishart matrix and identity matrix. In this paper, we generalize this decomposition to the polynomial high-dimensional regime $n \\asymp d^\\ell,\\ell \\in \\mathbb{N}$, for data uniformly distri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.10425","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/2204.10425/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":"2204.10425","created_at":"2026-07-05T04:16:56.731431+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.10425v1","created_at":"2026-07-05T04:16:56.731431+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.10425","created_at":"2026-07-05T04:16:56.731431+00:00"},{"alias_kind":"pith_short_12","alias_value":"XH42SYFBYF3M","created_at":"2026-07-05T04:16:56.731431+00:00"},{"alias_kind":"pith_short_16","alias_value":"XH42SYFBYF3MDDJM","created_at":"2026-07-05T04:16:56.731431+00:00"},{"alias_kind":"pith_short_8","alias_value":"XH42SYFB","created_at":"2026-07-05T04:16:56.731431+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27142","citing_title":"Estimation of High Dimensional Bounded Discrete Graphical Models via Regularized Generalized Score Matching","ref_index":176,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17767","citing_title":"Feature Learning in Linear-Width Two-Layer Networks: Two vs. One Step of Gradient Descent","ref_index":162,"is_internal_anchor":false},{"citing_arxiv_id":"2405.00592","citing_title":"Scaling and renormalization in high-dimensional regression","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17767","citing_title":"Feature Learning in Linear-Width Two-Layer Networks: Two vs. One Step of Gradient Descent","ref_index":162,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14524","citing_title":"Large Dimensional Kernel Ridge Regression: Extending to Product Kernels","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14567","citing_title":"Scaling Laws from Sequential Feature Recovery: A Solvable Hierarchical Model","ref_index":87,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I","json":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I.json","graph_json":"https://pith.science/api/pith-number/XH42SYFBYF3MDDJMFO6QFRCU7I/graph.json","events_json":"https://pith.science/api/pith-number/XH42SYFBYF3MDDJMFO6QFRCU7I/events.json","paper":"https://pith.science/paper/XH42SYFB"},"agent_actions":{"view_html":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I","download_json":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I.json","view_paper":"https://pith.science/paper/XH42SYFB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.10425&json=true","fetch_graph":"https://pith.science/api/pith-number/XH42SYFBYF3MDDJMFO6QFRCU7I/graph.json","fetch_events":"https://pith.science/api/pith-number/XH42SYFBYF3MDDJMFO6QFRCU7I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I/action/storage_attestation","attest_author":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I/action/author_attestation","sign_citation":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I/action/citation_signature","submit_replication":"https://pith.science/pith/XH42SYFBYF3MDDJMFO6QFRCU7I/action/replication_record"}},"created_at":"2026-07-05T04:16:56.731431+00:00","updated_at":"2026-07-05T04:16:56.731431+00:00"}