{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:DFT36U76PE4N36IPN2KIPJNJJF","short_pith_number":"pith:DFT36U76","schema_version":"1.0","canonical_sha256":"1967bf53fe7938ddf90f6e9487a5a9494b793e6a860fa645dcfc7aaadc2ad0ef","source":{"kind":"arxiv","id":"2202.09889","version":2},"attestation_state":"computed","paper":{"title":"Memorize to Generalize: on the Necessity of Interpolation in High Dimensional Linear Regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Chen Cheng, John Duchi, Rohith Kuditipudi","submitted_at":"2022-02-20T18:51:45Z","abstract_excerpt":"We examine the necessity of interpolation in overparameterized models, that is, when achieving optimal predictive risk in machine learning problems requires (nearly) interpolating the training data. In particular, we consider simple overparameterized linear regression $y = X \\theta + w$ with random design $X \\in \\mathbb{R}^{n \\times d}$ under the proportional asymptotics $d/n \\to \\gamma \\in (1, \\infty)$. We precisely characterize how prediction (test) error necessarily scales with training error in this setting. An implication of this characterization is that as the label noise variance $\\sigm"},"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":"2202.09889","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2022-02-20T18:51:45Z","cross_cats_sorted":["cs.LG","math.ST","stat.TH"],"title_canon_sha256":"97b955cbff79c14c3b7aeaecb568a070481a583039eeb7ad7ccdfd45056bc996","abstract_canon_sha256":"c2df0f6b93324beccee1f4a92999b2499adb09af8131c3255ae0da3acab78600"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:32:14.774556Z","signature_b64":"xcPUi2n1PbwjHmClxD4ZRIWWdEGyGsTl951Mle0p2by0k96xNswrz/bMZ6GsHTka4OlHTCaux+VxT8egeIgaCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1967bf53fe7938ddf90f6e9487a5a9494b793e6a860fa645dcfc7aaadc2ad0ef","last_reissued_at":"2026-07-05T04:32:14.774070Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:32:14.774070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memorize to Generalize: on the Necessity of Interpolation in High Dimensional Linear Regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Chen Cheng, John Duchi, Rohith Kuditipudi","submitted_at":"2022-02-20T18:51:45Z","abstract_excerpt":"We examine the necessity of interpolation in overparameterized models, that is, when achieving optimal predictive risk in machine learning problems requires (nearly) interpolating the training data. In particular, we consider simple overparameterized linear regression $y = X \\theta + w$ with random design $X \\in \\mathbb{R}^{n \\times d}$ under the proportional asymptotics $d/n \\to \\gamma \\in (1, \\infty)$. We precisely characterize how prediction (test) error necessarily scales with training error in this setting. An implication of this characterization is that as the label noise variance $\\sigm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.09889","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/2202.09889/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":"2202.09889","created_at":"2026-07-05T04:32:14.774127+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.09889v2","created_at":"2026-07-05T04:32:14.774127+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.09889","created_at":"2026-07-05T04:32:14.774127+00:00"},{"alias_kind":"pith_short_12","alias_value":"DFT36U76PE4N","created_at":"2026-07-05T04:32:14.774127+00:00"},{"alias_kind":"pith_short_16","alias_value":"DFT36U76PE4N36IP","created_at":"2026-07-05T04:32:14.774127+00:00"},{"alias_kind":"pith_short_8","alias_value":"DFT36U76","created_at":"2026-07-05T04:32:14.774127+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF","json":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF.json","graph_json":"https://pith.science/api/pith-number/DFT36U76PE4N36IPN2KIPJNJJF/graph.json","events_json":"https://pith.science/api/pith-number/DFT36U76PE4N36IPN2KIPJNJJF/events.json","paper":"https://pith.science/paper/DFT36U76"},"agent_actions":{"view_html":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF","download_json":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF.json","view_paper":"https://pith.science/paper/DFT36U76","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.09889&json=true","fetch_graph":"https://pith.science/api/pith-number/DFT36U76PE4N36IPN2KIPJNJJF/graph.json","fetch_events":"https://pith.science/api/pith-number/DFT36U76PE4N36IPN2KIPJNJJF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF/action/storage_attestation","attest_author":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF/action/author_attestation","sign_citation":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF/action/citation_signature","submit_replication":"https://pith.science/pith/DFT36U76PE4N36IPN2KIPJNJJF/action/replication_record"}},"created_at":"2026-07-05T04:32:14.774127+00:00","updated_at":"2026-07-05T04:32:14.774127+00:00"}