{"id":"W3136718240","doi":"10.1109/tcyb.2021.3059631","title":"Hierarchical Semantic Risk Minimization for Large-Scale Classification","year":2021,"lang":"en","type":"article","venue":"IEEE Transactions on Cybernetics","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Tianjin Science and Technology Program; National Natural Science Foundation of China","keywords":"Computer science; Hierarchy; Construct (python library); Artificial intelligence; Machine learning; Node (physics); Decision tree; Class hierarchy; Data mining; Class (philosophy); Task (project management); Scale (ratio); Tree (set theory); Mathematics","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":[0.0002137713,0.000143424,0.0001421396,0.0001120509,0.000348557,0.0001639661,0.0003221899,0.000125525,0.00002815945],"category_scores_gemma":[0.00005347391,0.0001509716,0.0001092406,0.0004390944,0.00002772627,0.0001886647,0.000003457208,0.0003062865,0.00006316358],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003649052,"about_ca_system_score_gemma":0.0001008474,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008363695,"about_ca_topic_score_gemma":0.00007579105,"domain_scores_codex":[0.9984871,0.0001883055,0.0002926147,0.0005204341,0.0002591227,0.0002524944],"domain_scores_gemma":[0.9985759,0.000308616,0.0001223765,0.0007445203,0.0001450716,0.0001035222],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001348293,0.003359059,0.001228601,0.0001883529,0.0001808307,0.00001151608,0.004646822,0.07143977,0.01863248,0.09230787,0.003821342,0.8040485],"study_design_scores_gemma":[0.0006383469,0.0001013713,0.004382799,0.00001903201,0.00005016028,0.0000114208,0.00007641176,0.9637234,0.01301018,0.001274632,0.01650587,0.0002063405],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003773023,0.00003083167,0.9920723,0.002392312,0.0005904344,0.0002009458,0.00009166762,0.0002097559,0.0006387597],"genre_scores_gemma":[0.8745378,0.0001992321,0.1236851,0.0002189122,0.00006706084,0.00006874782,0.00008834392,0.00001978981,0.001115019],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8922837,"threshold_uncertainty_score":0.6156442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02084593205790407,"score_gpt":0.2701675068500572,"score_spread":0.2493215747921532,"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."}}