{"id":"W4385849703","doi":"10.5281/zenodo.8252499","title":"Supplementary Information for: Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)","year":2023,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Vector Institute","funders":"","keywords":"Training (meteorology); Computer science; Machine learning; Artificial intelligence; Training set; 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":["sts","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.002496615,0.0001665692,0.0001961597,0.0005119491,0.006922427,0.0001940394,0.001072754,0.0001050897,0.009174242],"category_scores_gemma":[0.002312867,0.0001638244,0.00002878534,0.0008943525,0.0001028798,0.0007880134,0.001571286,0.0007558442,0.004320713],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002242618,"about_ca_system_score_gemma":0.00003741264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006328764,"about_ca_topic_score_gemma":0.00002953,"domain_scores_codex":[0.9971948,0.0005924312,0.0006465976,0.0003599129,0.0004240822,0.0007821714],"domain_scores_gemma":[0.9978817,0.0002925113,0.0002723536,0.0006162614,0.0007245535,0.0002126118],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0007149195,0.00006475709,0.0008027002,0.001015371,0.00009919857,0.0000109208,0.1239446,0.001724859,0.001089676,0.002988776,0.3061482,0.561396],"study_design_scores_gemma":[0.0004266513,0.0003075795,0.0003507641,0.0001544479,0.00001531053,0.00001170096,0.04111476,0.08740313,0.00006651495,0.00006069635,0.8699157,0.0001728003],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7639334,0.0001415866,0.07594541,0.02662562,0.002185094,0.01468117,0.04111674,0.01038714,0.06498383],"genre_scores_gemma":[0.9157096,0.00007660586,0.002286386,0.0004946276,0.00046646,0.000001297458,0.07869452,0.001186426,0.001084116],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5637674,"threshold_uncertainty_score":0.9964545,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3082396181610623,"score_gpt":0.4235564814636357,"score_spread":0.1153168633025735,"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."}}