{"id":"W3035618017","doi":"","title":"Improving Transformer Optimization Through Better Initialization","year":2020,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Power Transformer Diagnostics and Insulation","field":"Engineering","cited_by":51,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Initialization; Computer science; Transformer; Electrical engineering; Engineering; Voltage; Programming language","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00004875409,0.0001501152,0.0001077412,0.000060238,0.00006717869,0.0001085052,0.0001395621,0.00006462472,0.001017244],"category_scores_gemma":[0.00006570984,0.0001567958,0.00004533407,0.0001105192,0.00001528671,0.0004287148,0.000006846941,0.0003015642,0.00006613775],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003878603,"about_ca_system_score_gemma":0.00001368228,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002585009,"about_ca_topic_score_gemma":0.000003703167,"domain_scores_codex":[0.9991891,0.00002138141,0.0002330087,0.0001739265,0.00023777,0.0001447949],"domain_scores_gemma":[0.9997359,0.00003010635,0.00004011228,0.00004915905,0.00009016643,0.00005456432],"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":[0.00002713969,0.00001367687,0.0009935688,0.00002661266,0.00003923261,0.000003341546,0.0009314829,0.9659685,0.005424358,0.009679059,0.00005809329,0.0168349],"study_design_scores_gemma":[0.0004083718,0.00007649721,0.0002053948,0.00003124253,0.0000105318,0.000001291407,0.0000360806,0.991457,0.002320803,0.0002319934,0.005045383,0.0001753883],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007339437,0.00003202155,0.9323246,0.002620829,0.0003866554,0.0001208435,0.00002531842,0.0003092633,0.056841],"genre_scores_gemma":[0.9949431,0.0002315327,0.003210814,0.0009444894,0.0001934579,0.00001230178,0.000407455,0.00003459113,0.00002219915],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9876037,"threshold_uncertainty_score":0.999896,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02893562775711362,"score_gpt":0.2536630670916855,"score_spread":0.2247274393345718,"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."}}