{"id":"W3110846353","doi":"10.1609/aaai.v35i16.17680","title":"Reinforced Multi-Teacher Selection for Knowledge Distillation","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":121,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Bottleneck; Distillation; Computer science; Artificial intelligence; Inference; Machine learning; Selection (genetic algorithm); Learning cycle; Natural language processing; Mathematics education; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003462805,0.0001472424,0.0001729564,0.00006720752,0.000188637,0.0001914521,0.0007669693,0.00008719387,0.00002961366],"category_scores_gemma":[0.0007373811,0.0001216978,0.0001116919,0.0004792538,0.00006271769,0.0003357686,0.0002072787,0.0001679542,0.00002694428],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006146718,"about_ca_system_score_gemma":0.0001348534,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001386909,"about_ca_topic_score_gemma":0.00001678773,"domain_scores_codex":[0.9986849,0.00001330809,0.0004127607,0.0004316765,0.0002163888,0.0002409185],"domain_scores_gemma":[0.9983043,0.0000788015,0.0002299749,0.000210999,0.001122023,0.00005389818],"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":[0.00001754645,0.00005810908,0.000143966,0.00003988578,0.000009140077,5.042872e-8,0.001075811,0.0004751025,0.06448464,0.8663863,0.00005513068,0.06725433],"study_design_scores_gemma":[0.00002398577,0.0000366949,0.00005230567,0.00006201561,0.000005627083,0.000002027335,0.0001426165,0.5958485,0.3832543,0.02031871,0.0001574948,0.00009571602],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0561375,0.00002415144,0.9364303,0.001489707,0.0004947946,0.000376783,0.000001624914,0.0001017207,0.004943405],"genre_scores_gemma":[0.9700843,0.000008704552,0.02762394,0.00005221798,0.00009778247,0.00003665793,7.376495e-7,0.000009324704,0.002086348],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9139468,"threshold_uncertainty_score":0.4962693,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1365034617350052,"score_gpt":0.3305311318721846,"score_spread":0.1940276701371794,"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."}}