{"id":"W4287064118","doi":"10.48550/arxiv.2107.14194","title":"On the combined effect of class imbalance and concept complexity in deep\\n learning","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada; University of Alberta","funders":"","keywords":"Artificial intelligence; Overfitting; MNIST database; Machine learning; Deep learning; Computer science; Class (philosophy); Scarcity; Simple (philosophy); Artificial neural network; Economics","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":[0.0005212304,0.0002340002,0.0004110297,0.0001470072,0.0000977879,0.00008655265,0.001397849,0.000180053,0.0000132014],"category_scores_gemma":[0.0002225994,0.0002117249,0.00008470895,0.0005170048,0.0003474894,0.0001915711,0.001552104,0.000835785,0.000003965277],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001283543,"about_ca_system_score_gemma":0.00006035438,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009670288,"about_ca_topic_score_gemma":0.00003811917,"domain_scores_codex":[0.9979539,0.0007427098,0.0002086621,0.0007768021,0.000102827,0.0002150901],"domain_scores_gemma":[0.9975246,0.0008592352,0.0003236567,0.001141571,0.00009671414,0.00005426855],"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":[0.00009050479,0.0001087654,0.03060634,0.0001609424,0.00005761907,0.0001120841,0.0005319847,0.01779928,0.0005654879,0.9487555,0.0001375839,0.001073893],"study_design_scores_gemma":[0.001182456,0.0004161251,0.05370693,0.0004360639,0.0000294453,0.000003914565,0.000142417,0.9018767,0.009043194,0.03254059,0.000101353,0.0005208071],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7133704,0.00005800703,0.2845507,0.0002043634,0.0001180883,0.0004056807,0.000009316129,0.0001653745,0.001118107],"genre_scores_gemma":[0.9989539,0.0001095469,0.0007114576,0.00007430396,0.000007129949,0.000002954891,0.00003109457,0.000009328073,0.0001002979],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9162149,"threshold_uncertainty_score":0.863389,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05920185141972074,"score_gpt":0.2093491840486144,"score_spread":0.1501473326288937,"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."}}