{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:LHNDOPKMF5NHQ2P3D5GLTKIQOD","short_pith_number":"pith:LHNDOPKM","schema_version":"1.0","canonical_sha256":"59da373d4c2f5a7869fb1f4cb9a91070f799025955b911a9dc7adf5e47c07542","source":{"kind":"arxiv","id":"2203.11103","version":1},"attestation_state":"computed","paper":{"title":"Diverse Counterfactual Explanations for Anomaly Detection in Time Series","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Cedric Archambeau, Deborah Sulem, Francois-Xavier Aubet, Jan Gasthaus, Krishnaram Kenthapadi, Michele Donini, Muhammad Bilal Zafar, Sanjiv Das, Tim januschowski","submitted_at":"2022-03-21T16:30:34Z","abstract_excerpt":"Data-driven methods that detect anomalies in times series data are ubiquitous in practice, but they are in general unable to provide helpful explanations for the predictions they make. In this work we propose a model-agnostic algorithm that generates counterfactual ensemble explanations for time series anomaly detection models. Our method generates a set of diverse counterfactual examples, i.e, multiple perturbed versions of the original time series that are not considered anomalous by the detection model. Since the magnitude of the perturbations is limited, these counterfactuals represent an "},"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":"2203.11103","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-21T16:30:34Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"f8de9e17a686e30bad88b013dcfbb66797539ae52fc894b32c54d19dcbc23e57","abstract_canon_sha256":"26d159f333edd31784e9add1257edcd62e00f7539b1f5974cddc9d6e1a19e078"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:07:00.695888Z","signature_b64":"ItQxCYEOE+JjEDlwfuWGrtO0dbGnXbHk4JKEJzhGpVaPA/ZZrfVW23W2fsoWz6jm5LvMBH04eVE7fj4Mx2rnDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59da373d4c2f5a7869fb1f4cb9a91070f799025955b911a9dc7adf5e47c07542","last_reissued_at":"2026-07-05T04:07:00.695408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:07:00.695408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diverse Counterfactual Explanations for Anomaly Detection in Time Series","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Cedric Archambeau, Deborah Sulem, Francois-Xavier Aubet, Jan Gasthaus, Krishnaram Kenthapadi, Michele Donini, Muhammad Bilal Zafar, Sanjiv Das, Tim januschowski","submitted_at":"2022-03-21T16:30:34Z","abstract_excerpt":"Data-driven methods that detect anomalies in times series data are ubiquitous in practice, but they are in general unable to provide helpful explanations for the predictions they make. In this work we propose a model-agnostic algorithm that generates counterfactual ensemble explanations for time series anomaly detection models. Our method generates a set of diverse counterfactual examples, i.e, multiple perturbed versions of the original time series that are not considered anomalous by the detection model. Since the magnitude of the perturbations is limited, these counterfactuals represent an "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.11103","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/2203.11103/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":"2203.11103","created_at":"2026-07-05T04:07:00.695466+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.11103v1","created_at":"2026-07-05T04:07:00.695466+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.11103","created_at":"2026-07-05T04:07:00.695466+00:00"},{"alias_kind":"pith_short_12","alias_value":"LHNDOPKMF5NH","created_at":"2026-07-05T04:07:00.695466+00:00"},{"alias_kind":"pith_short_16","alias_value":"LHNDOPKMF5NHQ2P3","created_at":"2026-07-05T04:07:00.695466+00:00"},{"alias_kind":"pith_short_8","alias_value":"LHNDOPKM","created_at":"2026-07-05T04:07:00.695466+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.02300","citing_title":"A Meta Reinforcement Learning Approach to Goals-Based Wealth Management","ref_index":279,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD","json":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD.json","graph_json":"https://pith.science/api/pith-number/LHNDOPKMF5NHQ2P3D5GLTKIQOD/graph.json","events_json":"https://pith.science/api/pith-number/LHNDOPKMF5NHQ2P3D5GLTKIQOD/events.json","paper":"https://pith.science/paper/LHNDOPKM"},"agent_actions":{"view_html":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD","download_json":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD.json","view_paper":"https://pith.science/paper/LHNDOPKM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.11103&json=true","fetch_graph":"https://pith.science/api/pith-number/LHNDOPKMF5NHQ2P3D5GLTKIQOD/graph.json","fetch_events":"https://pith.science/api/pith-number/LHNDOPKMF5NHQ2P3D5GLTKIQOD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD/action/storage_attestation","attest_author":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD/action/author_attestation","sign_citation":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD/action/citation_signature","submit_replication":"https://pith.science/pith/LHNDOPKMF5NHQ2P3D5GLTKIQOD/action/replication_record"}},"created_at":"2026-07-05T04:07:00.695466+00:00","updated_at":"2026-07-05T04:07:00.695466+00:00"}