{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NEEPNDGTBFIEJ4U4XRIGXWE7EI","short_pith_number":"pith:NEEPNDGT","schema_version":"1.0","canonical_sha256":"6908f68cd3095044f29cbc506bd89f2233172c589e1257d6718b8671f7b8cd85","source":{"kind":"arxiv","id":"2409.09781","version":2},"attestation_state":"computed","paper":{"title":"RandALO: Out-of-sample risk estimation in no time flat","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC","stat.CO","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Daniel LeJeune, Emmanuel J. Cand\\`es, Parth Nobel","submitted_at":"2024-09-15T16:10:03Z","abstract_excerpt":"Estimating out-of-sample risk for models trained on large high-dimensional datasets is an expensive but essential part of the machine learning process, enabling practitioners to optimally tune hyperparameters. Cross-validation (CV) serves as the de facto standard for risk estimation but poorly trades off high bias ($K$-fold CV) for computational cost (leave-one-out CV). We propose a randomized approximate leave-one-out (RandALO) risk estimator that is not only a consistent estimator of risk in high dimensions but also less computationally expensive than $K$-fold CV. We support our claims with "},"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":"2409.09781","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2024-09-15T16:10:03Z","cross_cats_sorted":["cs.LG","math.OC","stat.CO","stat.ML","stat.TH"],"title_canon_sha256":"6ba3f5daf6c8a92fb765af8aaed0dce9d550428c854a44cc511b40cb1590026e","abstract_canon_sha256":"2437264183e31b97a90cb32efdfb1b81a26e133923afe8201efa093131f83c2b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:47.393610Z","signature_b64":"1MdyS/FUoDzbzRRPBsuxSsqR2/sSh8weHNdX8hR9b5ADSCdifT5LUNQ0mXHW/tmqJlZmfCqqnAfsmRh4w3cpBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6908f68cd3095044f29cbc506bd89f2233172c589e1257d6718b8671f7b8cd85","last_reissued_at":"2026-07-05T10:53:47.393043Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:47.393043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RandALO: Out-of-sample risk estimation in no time flat","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC","stat.CO","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Daniel LeJeune, Emmanuel J. Cand\\`es, Parth Nobel","submitted_at":"2024-09-15T16:10:03Z","abstract_excerpt":"Estimating out-of-sample risk for models trained on large high-dimensional datasets is an expensive but essential part of the machine learning process, enabling practitioners to optimally tune hyperparameters. Cross-validation (CV) serves as the de facto standard for risk estimation but poorly trades off high bias ($K$-fold CV) for computational cost (leave-one-out CV). We propose a randomized approximate leave-one-out (RandALO) risk estimator that is not only a consistent estimator of risk in high dimensions but also less computationally expensive than $K$-fold CV. We support our claims with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.09781","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/2409.09781/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":"2409.09781","created_at":"2026-07-05T10:53:47.393106+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.09781v2","created_at":"2026-07-05T10:53:47.393106+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.09781","created_at":"2026-07-05T10:53:47.393106+00:00"},{"alias_kind":"pith_short_12","alias_value":"NEEPNDGTBFIE","created_at":"2026-07-05T10:53:47.393106+00:00"},{"alias_kind":"pith_short_16","alias_value":"NEEPNDGTBFIEJ4U4","created_at":"2026-07-05T10:53:47.393106+00:00"},{"alias_kind":"pith_short_8","alias_value":"NEEPNDGT","created_at":"2026-07-05T10:53:47.393106+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.16430","citing_title":"MAGIC: Near-Optimal Data Attribution for Deep Learning","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI","json":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI.json","graph_json":"https://pith.science/api/pith-number/NEEPNDGTBFIEJ4U4XRIGXWE7EI/graph.json","events_json":"https://pith.science/api/pith-number/NEEPNDGTBFIEJ4U4XRIGXWE7EI/events.json","paper":"https://pith.science/paper/NEEPNDGT"},"agent_actions":{"view_html":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI","download_json":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI.json","view_paper":"https://pith.science/paper/NEEPNDGT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.09781&json=true","fetch_graph":"https://pith.science/api/pith-number/NEEPNDGTBFIEJ4U4XRIGXWE7EI/graph.json","fetch_events":"https://pith.science/api/pith-number/NEEPNDGTBFIEJ4U4XRIGXWE7EI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI/action/storage_attestation","attest_author":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI/action/author_attestation","sign_citation":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI/action/citation_signature","submit_replication":"https://pith.science/pith/NEEPNDGTBFIEJ4U4XRIGXWE7EI/action/replication_record"}},"created_at":"2026-07-05T10:53:47.393106+00:00","updated_at":"2026-07-05T10:53:47.393106+00:00"}