{"id":"W6966340950","doi":"10.48448/1h13-vc36","title":"Minimax and Neyman–Pearson Meta-Learning for Outlier Languages","year":2021,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Minimax; Outlier; Robustness (evolution); Bayes' theorem; Grammaticality; Differentiable function","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001462673,0.0004947206,0.0007945891,0.0009506138,0.000324916,0.0003978391,0.0006586105,0.0002439228,0.003264167],"category_scores_gemma":[0.0009461169,0.0004155819,0.000173735,0.000984277,0.001189084,0.0001672695,0.000354831,0.000403754,0.0003376169],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001156181,"about_ca_system_score_gemma":0.0004673014,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003274075,"about_ca_topic_score_gemma":0.0007976175,"domain_scores_codex":[0.9966785,0.0001036589,0.0003121672,0.001258247,0.0008602439,0.0007871524],"domain_scores_gemma":[0.9983846,0.0001892196,0.0003805153,0.0005883884,0.0001943348,0.0002629336],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00005871033,0.0004375285,0.001009312,0.0009840595,0.002741723,0.000188939,0.002327074,0.0002056003,0.04089034,0.008893983,0.8870786,0.05518414],"study_design_scores_gemma":[0.0006040747,0.0001359658,0.00011654,0.0001510903,0.0009945316,0.00003560505,0.001492727,0.004726067,0.0006320282,0.0001327408,0.9902431,0.0007355119],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.001240559,0.04233468,0.004164094,0.001175281,0.001198407,0.002553906,0.000606959,0.001695152,0.945031],"genre_scores_gemma":[0.007322493,0.0001376055,0.05778647,0.0003152601,0.000672186,0.0001102735,0.0002735386,0.001142798,0.9322394],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.1031645,"threshold_uncertainty_score":0.9998296,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05878942211465152,"score_gpt":0.3538531757982717,"score_spread":0.2950637536836203,"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."}}