{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TYI3PI7PBXAHRZKGPKFVYKSY2C","short_pith_number":"pith:TYI3PI7P","schema_version":"1.0","canonical_sha256":"9e11b7a3ef0dc078e5467a8b5c2a58d094b3aefde3896419883942614c770fcb","source":{"kind":"arxiv","id":"2107.14317","version":2},"attestation_state":"computed","paper":{"title":"Temporal Dependencies in Feature Importance for Time Series Predictions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Clayton Rooke, Jonathan Smith, Kin Kwan Leung, Maksims Volkovs, Saba Zuberi","submitted_at":"2021-07-29T20:31:03Z","abstract_excerpt":"Time series data introduces two key challenges for explainability methods: firstly, observations of the same feature over subsequent time steps are not independent, and secondly, the same feature can have varying importance to model predictions over time. In this paper, we propose Windowed Feature Importance in Time (WinIT), a feature removal based explainability approach to address these issues. Unlike existing feature removal explanation methods, WinIT explicitly accounts for the temporal dependence between different observations of the same feature in the construction of its importance scor"},"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":"2107.14317","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-07-29T20:31:03Z","cross_cats_sorted":[],"title_canon_sha256":"75a6b34ac7e76a69a8fe6cc23004ebf2b8cb80b97531f313aa498fce4bc53b54","abstract_canon_sha256":"606c078588126233c02c33ef72ee6d7cb40cd227e81af796ada610b27d80cd17"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:31.008296Z","signature_b64":"bLRQr1mA23WoCe89Qw+t7utrctTfxABfLgtezPYvNNiplc7WCPDLH2O1y/owwcKPMjQuhKZNUrURxriMXzDVAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e11b7a3ef0dc078e5467a8b5c2a58d094b3aefde3896419883942614c770fcb","last_reissued_at":"2026-07-05T05:48:31.007831Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:31.007831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Temporal Dependencies in Feature Importance for Time Series Predictions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Clayton Rooke, Jonathan Smith, Kin Kwan Leung, Maksims Volkovs, Saba Zuberi","submitted_at":"2021-07-29T20:31:03Z","abstract_excerpt":"Time series data introduces two key challenges for explainability methods: firstly, observations of the same feature over subsequent time steps are not independent, and secondly, the same feature can have varying importance to model predictions over time. In this paper, we propose Windowed Feature Importance in Time (WinIT), a feature removal based explainability approach to address these issues. Unlike existing feature removal explanation methods, WinIT explicitly accounts for the temporal dependence between different observations of the same feature in the construction of its importance scor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.14317","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/2107.14317/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":"2107.14317","created_at":"2026-07-05T05:48:31.007888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.14317v2","created_at":"2026-07-05T05:48:31.007888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.14317","created_at":"2026-07-05T05:48:31.007888+00:00"},{"alias_kind":"pith_short_12","alias_value":"TYI3PI7PBXAH","created_at":"2026-07-05T05:48:31.007888+00:00"},{"alias_kind":"pith_short_16","alias_value":"TYI3PI7PBXAHRZKG","created_at":"2026-07-05T05:48:31.007888+00:00"},{"alias_kind":"pith_short_8","alias_value":"TYI3PI7P","created_at":"2026-07-05T05:48:31.007888+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27599","citing_title":"Global Explanations for Multivariate Time Series Forecasting Models via $K$-Order Markov Approximations","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2511.23036","citing_title":"Delta-XAI: A Unified Framework for Explaining Prediction Changes in Online Time Series Monitoring","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C","json":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C.json","graph_json":"https://pith.science/api/pith-number/TYI3PI7PBXAHRZKGPKFVYKSY2C/graph.json","events_json":"https://pith.science/api/pith-number/TYI3PI7PBXAHRZKGPKFVYKSY2C/events.json","paper":"https://pith.science/paper/TYI3PI7P"},"agent_actions":{"view_html":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C","download_json":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C.json","view_paper":"https://pith.science/paper/TYI3PI7P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.14317&json=true","fetch_graph":"https://pith.science/api/pith-number/TYI3PI7PBXAHRZKGPKFVYKSY2C/graph.json","fetch_events":"https://pith.science/api/pith-number/TYI3PI7PBXAHRZKGPKFVYKSY2C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C/action/storage_attestation","attest_author":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C/action/author_attestation","sign_citation":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C/action/citation_signature","submit_replication":"https://pith.science/pith/TYI3PI7PBXAHRZKGPKFVYKSY2C/action/replication_record"}},"created_at":"2026-07-05T05:48:31.007888+00:00","updated_at":"2026-07-05T05:48:31.007888+00:00"}