{"id":"W2975674317","doi":"10.48550/arxiv.1909.11722","title":"A Theoretical Analysis of the Number of Shots in Few-Shot Learning","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Samsung","keywords":"Shot (pellet); Computer science; Hyperparameter; Task (project management); Set (abstract data type); Meta learning (computer science); Artificial intelligence; Machine learning; One shot; Simple (philosophy); Test (biology); Pattern recognition (psychology); Engineering","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":[0.0005501138,0.0001868056,0.0005010776,0.000426974,0.00005116698,0.0000343926,0.001614974,0.0002034188,0.0003027537],"category_scores_gemma":[0.0001315542,0.0001703089,0.0004368656,0.002324091,0.0002578645,0.0001297758,0.001496997,0.0007406176,0.00002311822],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008396197,"about_ca_system_score_gemma":0.00016009,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001172227,"about_ca_topic_score_gemma":0.00003028088,"domain_scores_codex":[0.9982126,0.0004677369,0.0003054104,0.0006139271,0.000172087,0.0002282505],"domain_scores_gemma":[0.9980841,0.0003146902,0.0004373572,0.0009641056,0.0001392468,0.00006053247],"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":[0.00001205056,0.00003447931,0.1090772,0.0000208205,0.0001486924,0.00000804163,0.0005131321,0.4503148,0.00002339837,0.4397536,0.0000023,0.00009148968],"study_design_scores_gemma":[0.0002881257,0.0000133974,0.07749225,0.00008107973,0.0002265263,8.009234e-7,0.0002225987,0.9074675,0.00006344415,0.01390142,0.00005745785,0.0001854207],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7633077,0.00001169218,0.2166088,0.00005562332,0.0001820029,0.000152065,0.000003460011,0.00003469256,0.01964399],"genre_scores_gemma":[0.9985886,0.00002346796,0.0005961888,0.00003846655,0.0000071474,3.237694e-7,0.000004890209,0.000008862719,0.0007320783],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4571527,"threshold_uncertainty_score":0.6944995,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05509535242507704,"score_gpt":0.2129186334918856,"score_spread":0.1578232810668085,"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."}}