{"id":"W3037881973","doi":"","title":"RelatIF: Identifying Explanatory Training Samples via Relative Influence","year":2020,"lang":"en","type":"article","venue":"International Conference on Artificial Intelligence and Statistics","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of Waterloo","funders":"","keywords":"Outlier; Computer science; Machine learning; Constraint (computer-aided design); Class (philosophy); Training (meteorology); Artificial intelligence; Econometrics; Mathematics; Geography","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":[],"consensus_categories":[],"category_scores_codex":[0.0002928786,0.0001667156,0.0001583408,0.000108559,0.0001913696,0.0004154711,0.0006455621,0.00006494222,0.0001230538],"category_scores_gemma":[0.001215817,0.0001678694,0.00002657004,0.0001803732,0.0001313204,0.0007023285,0.0001380579,0.000348748,0.0002393553],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000308419,"about_ca_system_score_gemma":0.0000846248,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005361393,"about_ca_topic_score_gemma":0.00002924796,"domain_scores_codex":[0.9983414,0.0001214003,0.00043246,0.0004973377,0.0004142366,0.0001930954],"domain_scores_gemma":[0.9986769,0.0005056251,0.0002027601,0.0002042376,0.0002483103,0.0001621418],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000211328,0.00001449486,0.0001517396,0.000005881813,0.00001445579,0.00001285894,0.003399819,0.0003050629,0.001048617,0.7731393,0.00005510506,0.2218316],"study_design_scores_gemma":[0.00003577042,0.0001402641,0.001760643,0.00005210378,0.000007166439,0.000009859763,0.0009738897,0.659647,0.0007052916,0.3350254,0.001400925,0.000241697],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001513022,0.00002095187,0.9887015,0.005533937,0.000263662,0.00009048685,0.00008615806,0.0001081742,0.003682108],"genre_scores_gemma":[0.9146397,0.00009388525,0.08412224,0.0009129306,0.0000816258,0.000009782701,0.00006909127,0.000008229452,0.00006256165],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9131266,"threshold_uncertainty_score":0.6845514,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2680933203368275,"score_gpt":0.3703719702066749,"score_spread":0.1022786498698474,"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."}}