{"id":"W3035097696","doi":"10.24963/ijcai.2020/296","title":"Handling Black Swan Events in Deep Learning with Diversely Extrapolated Neural Networks","year":2020,"lang":"en","type":"article","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval; McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Artificial intelligence; Hindsight bias; Deep learning; Artificial neural network; Set (abstract data type); Machine learning; Function (biology); Interpretability; Recurrent neural network","routes":{"ca_aff":true,"ca_fund":true,"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":[0.004671814,0.001321704,0.001432056,0.0008371945,0.0008120415,0.001238092,0.002754312,0.002502377,0.0008390796],"category_scores_gemma":[0.01401987,0.0008363544,0.0008542817,0.0007157319,0.002175018,0.003360434,0.004441511,0.004337169,0.0002083796],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001414796,"about_ca_system_score_gemma":0.001039823,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002829257,"about_ca_topic_score_gemma":0.003784841,"domain_scores_codex":[0.9985904,0.000471578,0.00007785799,0.0002873326,0.0003929203,0.000179828],"domain_scores_gemma":[0.9949219,0.003179121,0.0005400875,0.0006107342,0.0005052243,0.0002428656],"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":[0.000129496,0.00005271527,0.00307202,0.00004446748,0.00007419701,0.0002685541,0.0001664656,0.945375,0.001843349,0.01133024,0.0008275235,0.03681609],"study_design_scores_gemma":[0.000002766787,0.00001621479,0.00008183359,0.00000628055,0.000003780765,0.00001776429,0.000007410876,0.9900526,0.0005567234,0.009139101,0.000111989,0.000003439187],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07870688,0.0004560872,0.9185747,0.0005863459,0.00003930799,0.00002790915,0.00006137048,0.0007312589,0.0008161284],"genre_scores_gemma":[0.8616854,0.0002800837,0.1350647,0.0004622937,0.00008276186,0.0001040813,0.000205044,0.0001727419,0.001942889],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004671814,"threshold_uncertainty_score":0.0247072,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01508044470271067,"score_gpt":0.2180856509858087,"score_spread":0.203005206283098,"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."}}