{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:L5W3LMYYEGHKBYQRXSJ5WLB4VF","short_pith_number":"pith:L5W3LMYY","schema_version":"1.0","canonical_sha256":"5f6db5b318218ea0e211bc93db2c3ca95c6162c48fee48b07c110d42ad1b9343","source":{"kind":"arxiv","id":"2011.03687","version":3},"attestation_state":"computed","paper":{"title":"When Optimizing $f$-divergence is Robust with Label Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jiaheng Wei, Yang Liu","submitted_at":"2020-11-07T04:31:33Z","abstract_excerpt":"We show when maximizing a properly defined $f$-divergence measure with respect to a classifier's predictions and the supervised labels is robust with label noise. Leveraging its variational form, we derive a nice decoupling property for a family of $f$-divergence measures when label noise presents, where the divergence is shown to be a linear combination of the variational difference defined on the clean distribution and a bias term introduced due to the noise. The above derivation helps us analyze the robustness of different $f$-divergence functions. With established robustness, this family o"},"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":"2011.03687","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-11-07T04:31:33Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c837d921bd5936f37549372fe1a7fd4f617569ce7b7e5a2a442737e2df797ea6","abstract_canon_sha256":"a0396804ef5a0059a64deb17ebd907400438ded60f35d7a18e112634ac053338"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:07:03.947040Z","signature_b64":"Sf7mIuhxHwGLSQVuuHgp00Jx6lHe568vC38CslFFg0rsP+2ajIYmYMD+69YRG5l/Qcr6KHT5torVXtrntVKFCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f6db5b318218ea0e211bc93db2c3ca95c6162c48fee48b07c110d42ad1b9343","last_reissued_at":"2026-07-05T03:07:03.946615Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:07:03.946615Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Optimizing $f$-divergence is Robust with Label Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jiaheng Wei, Yang Liu","submitted_at":"2020-11-07T04:31:33Z","abstract_excerpt":"We show when maximizing a properly defined $f$-divergence measure with respect to a classifier's predictions and the supervised labels is robust with label noise. Leveraging its variational form, we derive a nice decoupling property for a family of $f$-divergence measures when label noise presents, where the divergence is shown to be a linear combination of the variational difference defined on the clean distribution and a bias term introduced due to the noise. The above derivation helps us analyze the robustness of different $f$-divergence functions. With established robustness, this family o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.03687","kind":"arxiv","version":3},"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/2011.03687/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":"2011.03687","created_at":"2026-07-05T03:07:03.946671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.03687v3","created_at":"2026-07-05T03:07:03.946671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.03687","created_at":"2026-07-05T03:07:03.946671+00:00"},{"alias_kind":"pith_short_12","alias_value":"L5W3LMYYEGHK","created_at":"2026-07-05T03:07:03.946671+00:00"},{"alias_kind":"pith_short_16","alias_value":"L5W3LMYYEGHKBYQR","created_at":"2026-07-05T03:07:03.946671+00:00"},{"alias_kind":"pith_short_8","alias_value":"L5W3LMYY","created_at":"2026-07-05T03:07:03.946671+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2408.11338","citing_title":"Automatic Dataset Construction (ADC): Sample Collection, Data Curation, and Beyond","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF","json":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF.json","graph_json":"https://pith.science/api/pith-number/L5W3LMYYEGHKBYQRXSJ5WLB4VF/graph.json","events_json":"https://pith.science/api/pith-number/L5W3LMYYEGHKBYQRXSJ5WLB4VF/events.json","paper":"https://pith.science/paper/L5W3LMYY"},"agent_actions":{"view_html":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF","download_json":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF.json","view_paper":"https://pith.science/paper/L5W3LMYY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.03687&json=true","fetch_graph":"https://pith.science/api/pith-number/L5W3LMYYEGHKBYQRXSJ5WLB4VF/graph.json","fetch_events":"https://pith.science/api/pith-number/L5W3LMYYEGHKBYQRXSJ5WLB4VF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF/action/storage_attestation","attest_author":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF/action/author_attestation","sign_citation":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF/action/citation_signature","submit_replication":"https://pith.science/pith/L5W3LMYYEGHKBYQRXSJ5WLB4VF/action/replication_record"}},"created_at":"2026-07-05T03:07:03.946671+00:00","updated_at":"2026-07-05T03:07:03.946671+00:00"}