{"id":"W3035856201","doi":"10.1109/icpr48806.2021.9412303","title":"A generalizable saliency map-based interpretation of model outcome","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Interpretability; Artificial intelligence; Computer science; Salient; Outcome (game theory); Class (philosophy); Machine learning; Pixel; Property (philosophy); Artificial neural network; Perspective (graphical); Sensitivity (control systems); Exploit; Pattern recognition (psychology); Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005414857,0.0006934557,0.0003631839,0.0006258538,0.0001984304,0.0008062291,0.000914699,0.0006647916,0.002677147],"category_scores_gemma":[0.003757968,0.0002530029,0.000549601,0.0003148045,0.0007402168,0.00122137,0.001104588,0.0009933246,0.0002960553],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005463082,"about_ca_system_score_gemma":0.0004416546,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001517756,"about_ca_topic_score_gemma":0.001511797,"domain_scores_codex":[0.9997784,0.00004580698,0.000008281321,0.000068931,0.00007402604,0.00002459161],"domain_scores_gemma":[0.9994774,0.000189989,0.00007018881,0.0001258462,0.0001057289,0.00003086775],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004838554,0.0001416172,0.003706188,0.0003383062,0.0001446099,0.0008985632,0.001075833,0.4640269,0.1184369,0.09987096,0.006255118,0.3046213],"study_design_scores_gemma":[0.00001531585,0.00005437444,0.0009139131,0.00001266152,0.00001481635,0.0001208298,0.00003643877,0.9568896,0.008721423,0.03210366,0.001104477,0.00001238987],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03360274,0.0000900564,0.9629688,0.0002753863,0.00002694822,0.00004597671,0.0001225291,0.0008010562,0.002066572],"genre_scores_gemma":[0.8213498,0.0001228105,0.1763053,0.0001000721,0.00004205017,0.00006464541,0.0003168267,0.0001644117,0.001534121],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002677147,"threshold_uncertainty_score":0.008955956,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06865303010564977,"score_gpt":0.3259663453591399,"score_spread":0.2573133152534901,"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."}}