{"id":"W3100210912","doi":"","title":"What went wrong and when? Instance-wise feature importance for time-series black-box models","year":2020,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Black box; Computer science; Series (stratigraphy); Feature (linguistics); Time series; Artificial intelligence; Machine learning","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":["scholarly_communication"],"category_scores_codex":[0.0002508543,0.0002214077,0.000262987,0.00008252468,0.000328021,0.003007367,0.0005081297,0.0001054296,0.000002259748],"category_scores_gemma":[0.00007487495,0.0001996113,0.0000465161,0.0003447836,0.00007844336,0.03101776,0.0001193665,0.0001559199,0.0000368999],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006173403,"about_ca_system_score_gemma":0.00009636896,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007217142,"about_ca_topic_score_gemma":0.000004711445,"domain_scores_codex":[0.9983821,0.00003272133,0.000584995,0.0002945357,0.0003604891,0.0003451909],"domain_scores_gemma":[0.9987018,0.00003998518,0.0004369845,0.0002601378,0.0003952058,0.0001658724],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004593492,0.00008727206,0.0004893107,0.008891777,0.00007911778,0.00002536685,0.2407925,0.2474367,0.002760822,0.1369243,0.04894177,0.3131116],"study_design_scores_gemma":[0.000154523,0.0001090024,0.000007431807,0.0001934515,0.000005529763,0.00001811711,0.00340406,0.9788014,0.001826861,0.001679327,0.01355661,0.0002437403],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01885686,0.002712191,0.9582,0.01588393,0.0009736817,0.001541669,0.0000148193,0.0006606046,0.001156228],"genre_scores_gemma":[0.9839424,0.0001244937,0.0115676,0.003166738,0.0002363593,0.0001788954,0.00003353156,0.00002306305,0.0007268808],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9650856,"threshold_uncertainty_score":0.9980276,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03220889811413114,"score_gpt":0.2498583712261862,"score_spread":0.2176494731120551,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}