{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DAIT3Q3W7XL6G6KFQF4A46AQLG","short_pith_number":"pith:DAIT3Q3W","schema_version":"1.0","canonical_sha256":"18113dc376fdd7e3794581780e78105984dcc208d4ba3fa974296cc284f73ad0","source":{"kind":"arxiv","id":"2507.04362","version":1},"attestation_state":"computed","paper":{"title":"Information-theoretic Quantification of High-order Feature Effects in Classification Problems","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["physics.data-an","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chiara Bar\\`a, Ivan Lazic, Luca Faes, Marta Iovino, Niksa Jakovljevic, Sebastiano Stramaglia","submitted_at":"2025-07-06T11:50:30Z","abstract_excerpt":"Understanding the contribution of individual features in predictive models remains a central goal in interpretable machine learning, and while many model-agnostic methods exist to estimate feature importance, they often fall short in capturing high-order interactions and disentangling overlapping contributions. In this work, we present an information-theoretic extension of the High-order interactions for Feature importance (Hi-Fi) method, leveraging Conditional Mutual Information (CMI) estimated via a k-Nearest Neighbor (kNN) approach working on mixed discrete and continuous random variables. "},"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":"2507.04362","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-06T11:50:30Z","cross_cats_sorted":["physics.data-an","stat.ML"],"title_canon_sha256":"e8aec7a93c8d3fc4a00eac0914c541030511d5264038739624989365d1cebbae","abstract_canon_sha256":"3fe470a63fdc2328fbf6a4c3888961fb9b51e81ff4d931cd7f9b67226aed2baf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:46.483913Z","signature_b64":"XPY6E2r8VfM+5fLCbo7GwmQEGsL9wTwBWGjHen/WYyO1gOy9MIp5j54zKNZ6QGIaKOxOwoQnso7pb4E784odAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"18113dc376fdd7e3794581780e78105984dcc208d4ba3fa974296cc284f73ad0","last_reissued_at":"2026-07-05T11:32:46.483463Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:46.483463Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Information-theoretic Quantification of High-order Feature Effects in Classification Problems","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["physics.data-an","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chiara Bar\\`a, Ivan Lazic, Luca Faes, Marta Iovino, Niksa Jakovljevic, Sebastiano Stramaglia","submitted_at":"2025-07-06T11:50:30Z","abstract_excerpt":"Understanding the contribution of individual features in predictive models remains a central goal in interpretable machine learning, and while many model-agnostic methods exist to estimate feature importance, they often fall short in capturing high-order interactions and disentangling overlapping contributions. In this work, we present an information-theoretic extension of the High-order interactions for Feature importance (Hi-Fi) method, leveraging Conditional Mutual Information (CMI) estimated via a k-Nearest Neighbor (kNN) approach working on mixed discrete and continuous random variables. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.04362","kind":"arxiv","version":1},"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/2507.04362/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":"2507.04362","created_at":"2026-07-05T11:32:46.483521+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.04362v1","created_at":"2026-07-05T11:32:46.483521+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.04362","created_at":"2026-07-05T11:32:46.483521+00:00"},{"alias_kind":"pith_short_12","alias_value":"DAIT3Q3W7XL6","created_at":"2026-07-05T11:32:46.483521+00:00"},{"alias_kind":"pith_short_16","alias_value":"DAIT3Q3W7XL6G6KF","created_at":"2026-07-05T11:32:46.483521+00:00"},{"alias_kind":"pith_short_8","alias_value":"DAIT3Q3W","created_at":"2026-07-05T11:32:46.483521+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.04362","citing_title":"Information-theoretic Quantification of High-order Feature Effects in Classification Problems","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG","json":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG.json","graph_json":"https://pith.science/api/pith-number/DAIT3Q3W7XL6G6KFQF4A46AQLG/graph.json","events_json":"https://pith.science/api/pith-number/DAIT3Q3W7XL6G6KFQF4A46AQLG/events.json","paper":"https://pith.science/paper/DAIT3Q3W"},"agent_actions":{"view_html":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG","download_json":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG.json","view_paper":"https://pith.science/paper/DAIT3Q3W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.04362&json=true","fetch_graph":"https://pith.science/api/pith-number/DAIT3Q3W7XL6G6KFQF4A46AQLG/graph.json","fetch_events":"https://pith.science/api/pith-number/DAIT3Q3W7XL6G6KFQF4A46AQLG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG/action/storage_attestation","attest_author":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG/action/author_attestation","sign_citation":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG/action/citation_signature","submit_replication":"https://pith.science/pith/DAIT3Q3W7XL6G6KFQF4A46AQLG/action/replication_record"}},"created_at":"2026-07-05T11:32:46.483521+00:00","updated_at":"2026-07-05T11:32:46.483521+00:00"}