{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:YBEWM6MINOLIZJWOFO6CPNHZZY","short_pith_number":"pith:YBEWM6MI","schema_version":"1.0","canonical_sha256":"c0496679886b968ca6ce2bbc27b4f9ce1db99d13ca87cdb118249d3b85d3c4ed","source":{"kind":"arxiv","id":"1801.05398","version":3},"attestation_state":"computed","paper":{"title":"On the Direction of Discrimination: An Information-Theoretic Analysis of Disparate Impact in Machine Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.IT","stat.ML"],"primary_cat":"cs.IT","authors_text":"Berk Ustun, Flavio P. Calmon, Hao Wang","submitted_at":"2018-01-16T18:26:56Z","abstract_excerpt":"In the context of machine learning, disparate impact refers to a form of systematic discrimination whereby the output distribution of a model depends on the value of a sensitive attribute (e.g., race or gender). In this paper, we propose an information-theoretic framework to analyze the disparate impact of a binary classification model. We view the model as a fixed channel, and quantify disparate impact as the divergence in output distributions over two groups. Our aim is to find a correction function that can perturb the input distributions of each group to align their output distributions. W"},"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":"1801.05398","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IT","submitted_at":"2018-01-16T18:26:56Z","cross_cats_sorted":["cs.LG","math.IT","stat.ML"],"title_canon_sha256":"4ad90b6dbc03d178305f67bdbb00b1af97c5ca247163e164c81f994aa8cdeb7c","abstract_canon_sha256":"f29f86f38244c2218650f53c407a4054225708edc3b70fa65a03f35b2748fe7c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:16:14.039977Z","signature_b64":"8I3seO9J0bjE0kxx8LQE/+3mzqYSDcYf6lr95QhM/1wn2SAOx5q4dSc82gQwqlA3aM78wy52oFAIVK5936iFBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0496679886b968ca6ce2bbc27b4f9ce1db99d13ca87cdb118249d3b85d3c4ed","last_reissued_at":"2026-05-18T00:16:14.039184Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:16:14.039184Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Direction of Discrimination: An Information-Theoretic Analysis of Disparate Impact in Machine Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.IT","stat.ML"],"primary_cat":"cs.IT","authors_text":"Berk Ustun, Flavio P. Calmon, Hao Wang","submitted_at":"2018-01-16T18:26:56Z","abstract_excerpt":"In the context of machine learning, disparate impact refers to a form of systematic discrimination whereby the output distribution of a model depends on the value of a sensitive attribute (e.g., race or gender). In this paper, we propose an information-theoretic framework to analyze the disparate impact of a binary classification model. We view the model as a fixed channel, and quantify disparate impact as the divergence in output distributions over two groups. Our aim is to find a correction function that can perturb the input distributions of each group to align their output distributions. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1801.05398","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":""},"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":"1801.05398","created_at":"2026-05-18T00:16:14.039325+00:00"},{"alias_kind":"arxiv_version","alias_value":"1801.05398v3","created_at":"2026-05-18T00:16:14.039325+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1801.05398","created_at":"2026-05-18T00:16:14.039325+00:00"},{"alias_kind":"pith_short_12","alias_value":"YBEWM6MINOLI","created_at":"2026-05-18T12:33:04.347982+00:00"},{"alias_kind":"pith_short_16","alias_value":"YBEWM6MINOLIZJWO","created_at":"2026-05-18T12:33:04.347982+00:00"},{"alias_kind":"pith_short_8","alias_value":"YBEWM6MI","created_at":"2026-05-18T12:33:04.347982+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/YBEWM6MINOLIZJWOFO6CPNHZZY","json":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY.json","graph_json":"https://pith.science/api/pith-number/YBEWM6MINOLIZJWOFO6CPNHZZY/graph.json","events_json":"https://pith.science/api/pith-number/YBEWM6MINOLIZJWOFO6CPNHZZY/events.json","paper":"https://pith.science/paper/YBEWM6MI"},"agent_actions":{"view_html":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY","download_json":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY.json","view_paper":"https://pith.science/paper/YBEWM6MI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1801.05398&json=true","fetch_graph":"https://pith.science/api/pith-number/YBEWM6MINOLIZJWOFO6CPNHZZY/graph.json","fetch_events":"https://pith.science/api/pith-number/YBEWM6MINOLIZJWOFO6CPNHZZY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY/action/storage_attestation","attest_author":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY/action/author_attestation","sign_citation":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY/action/citation_signature","submit_replication":"https://pith.science/pith/YBEWM6MINOLIZJWOFO6CPNHZZY/action/replication_record"}},"created_at":"2026-05-18T00:16:14.039325+00:00","updated_at":"2026-05-18T00:16:14.039325+00:00"}