{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JB2SOXMJ6P2UFCTFCH2SA5C5LL","short_pith_number":"pith:JB2SOXMJ","schema_version":"1.0","canonical_sha256":"4875275d89f3f5428a6511f520745d5afea55b5706ed8e2940f2ba115e48693c","source":{"kind":"arxiv","id":"2108.12250","version":2},"attestation_state":"computed","paper":{"title":"A comparison of approaches to improve worst-case predictive model performance over patient subpopulations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"stat.ML","authors_text":"Agata Foryciarz, Haoran Zhang, Marzyeh Ghassemi, Nigam H. Shah, Stephen R. Pfohl, Yizhe Xu","submitted_at":"2021-08-27T13:10:00Z","abstract_excerpt":"Predictive models for clinical outcomes that are accurate on average in a patient population may underperform drastically for some subpopulations, potentially introducing or reinforcing inequities in care access and quality. Model training approaches that aim to maximize worst-case model performance across subpopulations, such as distributionally robust optimization (DRO), attempt to address this problem without introducing additional harms. We conduct a large-scale empirical study of DRO and several variations of standard learning procedures to identify approaches for model development and se"},"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":"2108.12250","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2021-08-27T13:10:00Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"956b518a3343adf13ef01ba5afc2cff3b26820b1d621c8ecadfb482007ebff95","abstract_canon_sha256":"c26869f666f2d26dc7f8829154f26ebdc1cd2058794373e60c20a2016b70f63e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:53:25.649038Z","signature_b64":"E8i5AxjTQUL2mzRZWzVHpH7T9paygQYTQh4xEnKpXjpC70ruov8CrpaXDwPtmbR0DJXJYCZ8Aoz9FwywpMWLBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4875275d89f3f5428a6511f520745d5afea55b5706ed8e2940f2ba115e48693c","last_reissued_at":"2026-07-05T03:53:25.648623Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:53:25.648623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A comparison of approaches to improve worst-case predictive model performance over patient subpopulations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"stat.ML","authors_text":"Agata Foryciarz, Haoran Zhang, Marzyeh Ghassemi, Nigam H. Shah, Stephen R. Pfohl, Yizhe Xu","submitted_at":"2021-08-27T13:10:00Z","abstract_excerpt":"Predictive models for clinical outcomes that are accurate on average in a patient population may underperform drastically for some subpopulations, potentially introducing or reinforcing inequities in care access and quality. Model training approaches that aim to maximize worst-case model performance across subpopulations, such as distributionally robust optimization (DRO), attempt to address this problem without introducing additional harms. We conduct a large-scale empirical study of DRO and several variations of standard learning procedures to identify approaches for model development and se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.12250","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/2108.12250/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":"2108.12250","created_at":"2026-07-05T03:53:25.648677+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.12250v2","created_at":"2026-07-05T03:53:25.648677+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.12250","created_at":"2026-07-05T03:53:25.648677+00:00"},{"alias_kind":"pith_short_12","alias_value":"JB2SOXMJ6P2U","created_at":"2026-07-05T03:53:25.648677+00:00"},{"alias_kind":"pith_short_16","alias_value":"JB2SOXMJ6P2UFCTF","created_at":"2026-07-05T03:53:25.648677+00:00"},{"alias_kind":"pith_short_8","alias_value":"JB2SOXMJ","created_at":"2026-07-05T03:53:25.648677+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/JB2SOXMJ6P2UFCTFCH2SA5C5LL","json":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL.json","graph_json":"https://pith.science/api/pith-number/JB2SOXMJ6P2UFCTFCH2SA5C5LL/graph.json","events_json":"https://pith.science/api/pith-number/JB2SOXMJ6P2UFCTFCH2SA5C5LL/events.json","paper":"https://pith.science/paper/JB2SOXMJ"},"agent_actions":{"view_html":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL","download_json":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL.json","view_paper":"https://pith.science/paper/JB2SOXMJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.12250&json=true","fetch_graph":"https://pith.science/api/pith-number/JB2SOXMJ6P2UFCTFCH2SA5C5LL/graph.json","fetch_events":"https://pith.science/api/pith-number/JB2SOXMJ6P2UFCTFCH2SA5C5LL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL/action/storage_attestation","attest_author":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL/action/author_attestation","sign_citation":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL/action/citation_signature","submit_replication":"https://pith.science/pith/JB2SOXMJ6P2UFCTFCH2SA5C5LL/action/replication_record"}},"created_at":"2026-07-05T03:53:25.648677+00:00","updated_at":"2026-07-05T03:53:25.648677+00:00"}