{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:RMTWZXCOYVIYKFCCVLIUE4PCZD","short_pith_number":"pith:RMTWZXCO","schema_version":"1.0","canonical_sha256":"8b276cdc4ec551851442aad14271e2c8df11a98d955d7c0f346a860620892318","source":{"kind":"arxiv","id":"1908.10292","version":2},"attestation_state":"computed","paper":{"title":"On the Multiple Descent of Minimum-Norm Interpolants and Restricted Lower Isometry of Kernels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Alexander Rakhlin, Tengyuan Liang, Xiyu Zhai","submitted_at":"2019-08-27T16:05:50Z","abstract_excerpt":"We study the risk of minimum-norm interpolants of data in Reproducing Kernel Hilbert Spaces. Our upper bounds on the risk are of a multiple-descent shape for the various scalings of $d = n^{\\alpha}$, $\\alpha\\in(0,1)$, for the input dimension $d$ and sample size $n$. Empirical evidence supports our finding that minimum-norm interpolants in RKHS can exhibit this unusual non-monotonicity in sample size; furthermore, locations of the peaks in our experiments match our theoretical predictions. Since gradient flow on appropriately initialized wide neural networks converges to a minimum-norm interpol"},"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":"1908.10292","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2019-08-27T16:05:50Z","cross_cats_sorted":["cs.LG","stat.ML","stat.TH"],"title_canon_sha256":"dfc4fa37010be6a9f40266afeba288a4a40aa313f187c6ac091833c74396f022","abstract_canon_sha256":"e438b2cac38d2d6ac4b2d6adc3d85cc294da83d8de346eb3a912581e3c6594a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:21:46.138389Z","signature_b64":"b4lJWfXzH6Q+cAk8fm7DDJUhiK1aMOxcQPNkOY27obXyJY/KCmhvBRO6L5aT1wvX4N5S72hZJ8YwSV4sxLypDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b276cdc4ec551851442aad14271e2c8df11a98d955d7c0f346a860620892318","last_reissued_at":"2026-07-05T01:21:46.137930Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:21:46.137930Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Multiple Descent of Minimum-Norm Interpolants and Restricted Lower Isometry of Kernels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Alexander Rakhlin, Tengyuan Liang, Xiyu Zhai","submitted_at":"2019-08-27T16:05:50Z","abstract_excerpt":"We study the risk of minimum-norm interpolants of data in Reproducing Kernel Hilbert Spaces. Our upper bounds on the risk are of a multiple-descent shape for the various scalings of $d = n^{\\alpha}$, $\\alpha\\in(0,1)$, for the input dimension $d$ and sample size $n$. Empirical evidence supports our finding that minimum-norm interpolants in RKHS can exhibit this unusual non-monotonicity in sample size; furthermore, locations of the peaks in our experiments match our theoretical predictions. Since gradient flow on appropriately initialized wide neural networks converges to a minimum-norm interpol"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.10292","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/1908.10292/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":"1908.10292","created_at":"2026-07-05T01:21:46.137992+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.10292v2","created_at":"2026-07-05T01:21:46.137992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.10292","created_at":"2026-07-05T01:21:46.137992+00:00"},{"alias_kind":"pith_short_12","alias_value":"RMTWZXCOYVIY","created_at":"2026-07-05T01:21:46.137992+00:00"},{"alias_kind":"pith_short_16","alias_value":"RMTWZXCOYVIYKFCC","created_at":"2026-07-05T01:21:46.137992+00:00"},{"alias_kind":"pith_short_8","alias_value":"RMTWZXCO","created_at":"2026-07-05T01:21:46.137992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.14757","citing_title":"Solving Inverse Parametrized Problems via Finite Elements and Extreme Learning Networks","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD","json":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD.json","graph_json":"https://pith.science/api/pith-number/RMTWZXCOYVIYKFCCVLIUE4PCZD/graph.json","events_json":"https://pith.science/api/pith-number/RMTWZXCOYVIYKFCCVLIUE4PCZD/events.json","paper":"https://pith.science/paper/RMTWZXCO"},"agent_actions":{"view_html":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD","download_json":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD.json","view_paper":"https://pith.science/paper/RMTWZXCO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.10292&json=true","fetch_graph":"https://pith.science/api/pith-number/RMTWZXCOYVIYKFCCVLIUE4PCZD/graph.json","fetch_events":"https://pith.science/api/pith-number/RMTWZXCOYVIYKFCCVLIUE4PCZD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD/action/storage_attestation","attest_author":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD/action/author_attestation","sign_citation":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD/action/citation_signature","submit_replication":"https://pith.science/pith/RMTWZXCOYVIYKFCCVLIUE4PCZD/action/replication_record"}},"created_at":"2026-07-05T01:21:46.137992+00:00","updated_at":"2026-07-05T01:21:46.137992+00:00"}