{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:WXF23UTGKEB55VCKPO2JTRN6AD","short_pith_number":"pith:WXF23UTG","schema_version":"1.0","canonical_sha256":"b5cbadd2665103ded44a7bb499c5be00deee3101b57f204fde42119aefbc706e","source":{"kind":"arxiv","id":"2007.01162","version":2},"attestation_state":"computed","paper":{"title":"Tilted Empirical Risk Minimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IT","math.IT","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ahmad Beirami, Maziar Sanjabi, Tian Li, Virginia Smith","submitted_at":"2020-07-02T14:49:48Z","abstract_excerpt":"Empirical risk minimization (ERM) is typically designed to perform well on the average loss, which can result in estimators that are sensitive to outliers, generalize poorly, or treat subgroups unfairly. While many methods aim to address these problems individually, in this work, we explore them through a unified framework -- tilted empirical risk minimization (TERM). In particular, we show that it is possible to flexibly tune the impact of individual losses through a straightforward extension to ERM using a hyperparameter called the tilt. We provide several interpretations of the resulting fr"},"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":"2007.01162","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-07-02T14:49:48Z","cross_cats_sorted":["cs.IT","math.IT","stat.ML"],"title_canon_sha256":"12c3c26edc59da9d96ac67ef6225aa0e485680cb2cbb18bbe5fe593d50f98da8","abstract_canon_sha256":"7b0adf9fad36b2bc839a9abc8f0d22991a34e64e5c3f9f93a0aa24e5c35d96f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:23:57.510428Z","signature_b64":"k/wGrzpcd5rKXTc8FKM2rwOB3xcQrvgYvuzBM+Ai4TvVDxQIQX9mBBLFMcNLssyjSzBKtu08FYmyIdnfqVdTDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b5cbadd2665103ded44a7bb499c5be00deee3101b57f204fde42119aefbc706e","last_reissued_at":"2026-07-05T02:23:57.509950Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:23:57.509950Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tilted Empirical Risk Minimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IT","math.IT","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ahmad Beirami, Maziar Sanjabi, Tian Li, Virginia Smith","submitted_at":"2020-07-02T14:49:48Z","abstract_excerpt":"Empirical risk minimization (ERM) is typically designed to perform well on the average loss, which can result in estimators that are sensitive to outliers, generalize poorly, or treat subgroups unfairly. While many methods aim to address these problems individually, in this work, we explore them through a unified framework -- tilted empirical risk minimization (TERM). In particular, we show that it is possible to flexibly tune the impact of individual losses through a straightforward extension to ERM using a hyperparameter called the tilt. We provide several interpretations of the resulting fr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.01162","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/2007.01162/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":"2007.01162","created_at":"2026-07-05T02:23:57.510004+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.01162v2","created_at":"2026-07-05T02:23:57.510004+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.01162","created_at":"2026-07-05T02:23:57.510004+00:00"},{"alias_kind":"pith_short_12","alias_value":"WXF23UTGKEB5","created_at":"2026-07-05T02:23:57.510004+00:00"},{"alias_kind":"pith_short_16","alias_value":"WXF23UTGKEB55VCK","created_at":"2026-07-05T02:23:57.510004+00:00"},{"alias_kind":"pith_short_8","alias_value":"WXF23UTG","created_at":"2026-07-05T02:23:57.510004+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01561","citing_title":"S-SPPO: Semantic-Calibrated Self-Play Preference Optimization","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21849","citing_title":"Geometry-Adaptive Explainer for Faithful Dictionary-Based Interpretability under Distribution Shift","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10521","citing_title":"DuetFair: Coupling Inter- and Intra-Subgroup Robustness for Fair Medical Image Segmentation","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD","json":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD.json","graph_json":"https://pith.science/api/pith-number/WXF23UTGKEB55VCKPO2JTRN6AD/graph.json","events_json":"https://pith.science/api/pith-number/WXF23UTGKEB55VCKPO2JTRN6AD/events.json","paper":"https://pith.science/paper/WXF23UTG"},"agent_actions":{"view_html":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD","download_json":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD.json","view_paper":"https://pith.science/paper/WXF23UTG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.01162&json=true","fetch_graph":"https://pith.science/api/pith-number/WXF23UTGKEB55VCKPO2JTRN6AD/graph.json","fetch_events":"https://pith.science/api/pith-number/WXF23UTGKEB55VCKPO2JTRN6AD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD/action/storage_attestation","attest_author":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD/action/author_attestation","sign_citation":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD/action/citation_signature","submit_replication":"https://pith.science/pith/WXF23UTGKEB55VCKPO2JTRN6AD/action/replication_record"}},"created_at":"2026-07-05T02:23:57.510004+00:00","updated_at":"2026-07-05T02:23:57.510004+00:00"}