{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TLUN5I5INURGXIWONEG3JEM76R","short_pith_number":"pith:TLUN5I5I","schema_version":"1.0","canonical_sha256":"9ae8dea3a86d226ba2ce690db4919ff4561aa0f1333e2e4d4b9aff034bbb0cc6","source":{"kind":"arxiv","id":"2106.05303","version":1},"attestation_state":"computed","paper":{"title":"Explaining Time Series Predictions with Dynamic Masks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jonathan Crabb\\'e, Mihaela van der Schaar","submitted_at":"2021-06-09T18:01:09Z","abstract_excerpt":"How can we explain the predictions of a machine learning model? When the data is structured as a multivariate time series, this question induces additional difficulties such as the necessity for the explanation to embody the time dependency and the large number of inputs. To address these challenges, we propose dynamic masks (Dynamask). This method produces instance-wise importance scores for each feature at each time step by fitting a perturbation mask to the input sequence. In order to incorporate the time dependency of the data, Dynamask studies the effects of dynamic perturbation operators"},"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":"2106.05303","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-09T18:01:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"220c33b9c13d04246754f9002da510e4bfc516d793dd3a86d053aab1e5ac428a","abstract_canon_sha256":"90890840ee6d16114e67ef978d73f8aadfbfa8164335c3b6c99a0b9d90ca6fa2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:48:11.686474Z","signature_b64":"vqRRfh0ig+rOBh/sMMmw5L8tvPVF4icEUsdMqN0zLHMlw1Tp1lrwGk4HuKo+D9C5/meRfJvZTXhu+L76vRAlDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ae8dea3a86d226ba2ce690db4919ff4561aa0f1333e2e4d4b9aff034bbb0cc6","last_reissued_at":"2026-07-05T02:48:11.686133Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:48:11.686133Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Explaining Time Series Predictions with Dynamic Masks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jonathan Crabb\\'e, Mihaela van der Schaar","submitted_at":"2021-06-09T18:01:09Z","abstract_excerpt":"How can we explain the predictions of a machine learning model? When the data is structured as a multivariate time series, this question induces additional difficulties such as the necessity for the explanation to embody the time dependency and the large number of inputs. To address these challenges, we propose dynamic masks (Dynamask). This method produces instance-wise importance scores for each feature at each time step by fitting a perturbation mask to the input sequence. In order to incorporate the time dependency of the data, Dynamask studies the effects of dynamic perturbation operators"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.05303","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/2106.05303/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":"2106.05303","created_at":"2026-07-05T02:48:11.686187+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.05303v1","created_at":"2026-07-05T02:48:11.686187+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.05303","created_at":"2026-07-05T02:48:11.686187+00:00"},{"alias_kind":"pith_short_12","alias_value":"TLUN5I5INURG","created_at":"2026-07-05T02:48:11.686187+00:00"},{"alias_kind":"pith_short_16","alias_value":"TLUN5I5INURGXIWO","created_at":"2026-07-05T02:48:11.686187+00:00"},{"alias_kind":"pith_short_8","alias_value":"TLUN5I5I","created_at":"2026-07-05T02:48:11.686187+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.17783","citing_title":"Feature Attribution-Based Explainability Analysis of Deep Learning Models in Predictive Process Monitoring","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R","json":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R.json","graph_json":"https://pith.science/api/pith-number/TLUN5I5INURGXIWONEG3JEM76R/graph.json","events_json":"https://pith.science/api/pith-number/TLUN5I5INURGXIWONEG3JEM76R/events.json","paper":"https://pith.science/paper/TLUN5I5I"},"agent_actions":{"view_html":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R","download_json":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R.json","view_paper":"https://pith.science/paper/TLUN5I5I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.05303&json=true","fetch_graph":"https://pith.science/api/pith-number/TLUN5I5INURGXIWONEG3JEM76R/graph.json","fetch_events":"https://pith.science/api/pith-number/TLUN5I5INURGXIWONEG3JEM76R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R/action/storage_attestation","attest_author":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R/action/author_attestation","sign_citation":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R/action/citation_signature","submit_replication":"https://pith.science/pith/TLUN5I5INURGXIWONEG3JEM76R/action/replication_record"}},"created_at":"2026-07-05T02:48:11.686187+00:00","updated_at":"2026-07-05T02:48:11.686187+00:00"}