{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:RMUNGK3JTEOKC6WBQ44OV4WLR5","short_pith_number":"pith:RMUNGK3J","schema_version":"1.0","canonical_sha256":"8b28d32b69991ca17ac18738eaf2cb8f4377823fc6e2674203fb5531066bfa49","source":{"kind":"arxiv","id":"1907.07165","version":2},"attestation_state":"computed","paper":{"title":"Explaining Classifiers with Causal Concept Effect (CaCE)","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Amir Feder, Been Kim, Uri Shalit, Yash Goyal","submitted_at":"2019-07-16T17:47:43Z","abstract_excerpt":"How can we understand classification decisions made by deep neural networks? Many existing explainability methods rely solely on correlations and fail to account for confounding, which may result in potentially misleading explanations. To overcome this problem, we define the Causal Concept Effect (CaCE) as the causal effect of (the presence or absence of) a human-interpretable concept on a deep neural net's predictions. We show that the CaCE measure can avoid errors stemming from confounding. Estimating CaCE is difficult in situations where we cannot easily simulate the do-operator. To mitigat"},"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":"1907.07165","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-16T17:47:43Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"75ebcb56d930744c3b0ba665bc740b7d0c988b53aec2a71b93a089ec8b74bc46","abstract_canon_sha256":"8a49bd56924828071076f0aa2f6374c880249fde523d791808ae289c6f3e62e3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:44:26.049024Z","signature_b64":"RoMP5aKBe6TSiHxpu8TA6Lczmyw65ynfFm47FjFL+cQGvv+ArUvqFRWTd76LN2gjVUc+H2u4m1g8ngT7o4qdAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b28d32b69991ca17ac18738eaf2cb8f4377823fc6e2674203fb5531066bfa49","last_reissued_at":"2026-07-05T00:44:26.048611Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:44:26.048611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Explaining Classifiers with Causal Concept Effect (CaCE)","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Amir Feder, Been Kim, Uri Shalit, Yash Goyal","submitted_at":"2019-07-16T17:47:43Z","abstract_excerpt":"How can we understand classification decisions made by deep neural networks? Many existing explainability methods rely solely on correlations and fail to account for confounding, which may result in potentially misleading explanations. To overcome this problem, we define the Causal Concept Effect (CaCE) as the causal effect of (the presence or absence of) a human-interpretable concept on a deep neural net's predictions. We show that the CaCE measure can avoid errors stemming from confounding. Estimating CaCE is difficult in situations where we cannot easily simulate the do-operator. To mitigat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.07165","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/1907.07165/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":"1907.07165","created_at":"2026-07-05T00:44:26.048664+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.07165v2","created_at":"2026-07-05T00:44:26.048664+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.07165","created_at":"2026-07-05T00:44:26.048664+00:00"},{"alias_kind":"pith_short_12","alias_value":"RMUNGK3JTEOK","created_at":"2026-07-05T00:44:26.048664+00:00"},{"alias_kind":"pith_short_16","alias_value":"RMUNGK3JTEOKC6WB","created_at":"2026-07-05T00:44:26.048664+00:00"},{"alias_kind":"pith_short_8","alias_value":"RMUNGK3J","created_at":"2026-07-05T00:44:26.048664+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30313","citing_title":"TRACE: A Concept Bottleneck Model for Longitudinal 3D Glioblastoma Response Assessment","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00267","citing_title":"Validating Causal Abstraction Metrics on Simulated Complex Systems","ref_index":174,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30313","citing_title":"TRACE: A Concept Bottleneck Model for Longitudinal 3D Glioblastoma Response Assessment","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05126","citing_title":"CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2504.11159","citing_title":"C-SHAP for time series: An approach to high-level temporal explanations","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05126","citing_title":"CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2601.16074","citing_title":"Explainable AI to Improve Machine Learning Reliability for Industrial Cyber-Physical Systems","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5","json":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5.json","graph_json":"https://pith.science/api/pith-number/RMUNGK3JTEOKC6WBQ44OV4WLR5/graph.json","events_json":"https://pith.science/api/pith-number/RMUNGK3JTEOKC6WBQ44OV4WLR5/events.json","paper":"https://pith.science/paper/RMUNGK3J"},"agent_actions":{"view_html":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5","download_json":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5.json","view_paper":"https://pith.science/paper/RMUNGK3J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.07165&json=true","fetch_graph":"https://pith.science/api/pith-number/RMUNGK3JTEOKC6WBQ44OV4WLR5/graph.json","fetch_events":"https://pith.science/api/pith-number/RMUNGK3JTEOKC6WBQ44OV4WLR5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5/action/storage_attestation","attest_author":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5/action/author_attestation","sign_citation":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5/action/citation_signature","submit_replication":"https://pith.science/pith/RMUNGK3JTEOKC6WBQ44OV4WLR5/action/replication_record"}},"created_at":"2026-07-05T00:44:26.048664+00:00","updated_at":"2026-07-05T00:44:26.048664+00:00"}