{"id":"W3099471123","doi":"10.48550/arxiv.2011.10036","title":"On the Dynamics of Training Attention Models","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Discriminative model; Computer science; Classifier (UML); Artificial intelligence; Embedding; Machine learning; Stochastic gradient descent; Dynamics (music); Task (project management); Set (abstract data type); Block (permutation group theory); Simple (philosophy); Artificial neural network; Natural language processing; Psychology; Mathematics; Engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002976885,0.0009338377,0.001244675,0.0007103471,0.0007124922,0.001474203,0.001342137,0.001977806,0.00428139],"category_scores_gemma":[0.03682711,0.0007745613,0.0004695363,0.0005319065,0.001876434,0.003978,0.002489391,0.003286424,0.0005419303],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001744692,"about_ca_system_score_gemma":0.001106979,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006202447,"about_ca_topic_score_gemma":0.005221822,"domain_scores_codex":[0.9991253,0.0003793338,0.00003328907,0.0002103412,0.0001267293,0.0001251168],"domain_scores_gemma":[0.9884365,0.009303534,0.0006385973,0.0005041118,0.0006934116,0.0004238626],"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":[0.000278362,0.0001402623,0.006372054,0.0002242627,0.00008633573,0.0002114424,0.0007204336,0.7128887,0.007372369,0.2119325,0.005375613,0.05439774],"study_design_scores_gemma":[0.00001146629,0.00002655137,0.0004066003,0.00001941612,0.000007713556,0.00002057835,0.00002430797,0.9573498,0.0003701599,0.04141607,0.0003407047,0.000006619611],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.224792,0.002006042,0.7534718,0.00597207,0.0001465035,0.0000962107,0.0002687943,0.0007533889,0.01249318],"genre_scores_gemma":[0.9553373,0.0007915043,0.03699519,0.0005615658,0.000111239,0.0001757781,0.0002754357,0.0002467006,0.005505173],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006202447,"threshold_uncertainty_score":0.01574343,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1867217059251652,"score_gpt":0.1889230455187379,"score_spread":0.002201339593572688,"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."}}