{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6FESK3URTKMNW2IR6AFYKIS4RB","short_pith_number":"pith:6FESK3UR","schema_version":"1.0","canonical_sha256":"f149256e919a98db6911f00b85225c8843601ccd7841e42d14b6a547aaca0a14","source":{"kind":"arxiv","id":"2404.18702","version":2},"attestation_state":"computed","paper":{"title":"Why You Should Not Trust Interpretations in Machine Learning: Adversarial Attacks on Partial Dependence Plots","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","stat.AP","stat.ML"],"primary_cat":"cs.LG","authors_text":"Fei Huang, Giles Hooker, Xi Xin","submitted_at":"2024-04-29T13:51:41Z","abstract_excerpt":"The adoption of artificial intelligence (AI) across industries has led to the widespread use of complex black-box models and interpretation tools for decision making. This paper proposes an adversarial framework to uncover the vulnerability of permutation-based interpretation methods for machine learning tasks, with a particular focus on partial dependence (PD) plots. This adversarial framework modifies the original black box model to manipulate its predictions for instances in the extrapolation domain. As a result, it produces deceptive PD plots that can conceal discriminatory behaviors while"},"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":"2404.18702","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-29T13:51:41Z","cross_cats_sorted":["cs.CR","stat.AP","stat.ML"],"title_canon_sha256":"1ceee92a45490d693d1cdfe8e8c3a08922eb234fd450d732303e9219b7ad6c2e","abstract_canon_sha256":"4a4859fcc64ddede9422598bfdc1f9d3311a71623254e63c07361201cace908b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:14:06.436307Z","signature_b64":"lhjE01MeIosNw0j7v1TtFTnHyDn4PCbMEDz7jfSJQOszHJM6LmDIrMLu82jymZ7+C7vcAxfcyPT7noadJERTAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f149256e919a98db6911f00b85225c8843601ccd7841e42d14b6a547aaca0a14","last_reissued_at":"2026-07-05T08:14:06.435838Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:14:06.435838Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why You Should Not Trust Interpretations in Machine Learning: Adversarial Attacks on Partial Dependence Plots","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","stat.AP","stat.ML"],"primary_cat":"cs.LG","authors_text":"Fei Huang, Giles Hooker, Xi Xin","submitted_at":"2024-04-29T13:51:41Z","abstract_excerpt":"The adoption of artificial intelligence (AI) across industries has led to the widespread use of complex black-box models and interpretation tools for decision making. This paper proposes an adversarial framework to uncover the vulnerability of permutation-based interpretation methods for machine learning tasks, with a particular focus on partial dependence (PD) plots. This adversarial framework modifies the original black box model to manipulate its predictions for instances in the extrapolation domain. As a result, it produces deceptive PD plots that can conceal discriminatory behaviors while"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.18702","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/2404.18702/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":"2404.18702","created_at":"2026-07-05T08:14:06.435900+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.18702v2","created_at":"2026-07-05T08:14:06.435900+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.18702","created_at":"2026-07-05T08:14:06.435900+00:00"},{"alias_kind":"pith_short_12","alias_value":"6FESK3URTKMN","created_at":"2026-07-05T08:14:06.435900+00:00"},{"alias_kind":"pith_short_16","alias_value":"6FESK3URTKMNW2IR","created_at":"2026-07-05T08:14:06.435900+00:00"},{"alias_kind":"pith_short_8","alias_value":"6FESK3UR","created_at":"2026-07-05T08:14:06.435900+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2409.14590","citing_title":"Explainable AI needs formalization","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB","json":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB.json","graph_json":"https://pith.science/api/pith-number/6FESK3URTKMNW2IR6AFYKIS4RB/graph.json","events_json":"https://pith.science/api/pith-number/6FESK3URTKMNW2IR6AFYKIS4RB/events.json","paper":"https://pith.science/paper/6FESK3UR"},"agent_actions":{"view_html":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB","download_json":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB.json","view_paper":"https://pith.science/paper/6FESK3UR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.18702&json=true","fetch_graph":"https://pith.science/api/pith-number/6FESK3URTKMNW2IR6AFYKIS4RB/graph.json","fetch_events":"https://pith.science/api/pith-number/6FESK3URTKMNW2IR6AFYKIS4RB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB/action/storage_attestation","attest_author":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB/action/author_attestation","sign_citation":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB/action/citation_signature","submit_replication":"https://pith.science/pith/6FESK3URTKMNW2IR6AFYKIS4RB/action/replication_record"}},"created_at":"2026-07-05T08:14:06.435900+00:00","updated_at":"2026-07-05T08:14:06.435900+00:00"}