{"id":"W2736761179","doi":"10.1016/j.patcog.2017.07.019","title":"On automated source selection for transfer learning in convolutional neural networks","year":2017,"lang":"en","type":"article","venue":"Pattern Recognition","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":127,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"National Institute on Aging; National Institutes of Health; National Institute of Diabetes and Digestive and Kidney Diseases; National Cancer Institute; EMS Ingénierie","keywords":"Transfer of learning; Computer science; Convolutional neural network; Artificial intelligence; Selection (genetic algorithm); Machine learning; Pattern recognition (psychology)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00323567,0.001466869,0.00222254,0.001208421,0.001010576,0.001355693,0.003662711,0.002579688,0.004830047],"category_scores_gemma":[0.01163823,0.0008981841,0.001180435,0.001222729,0.001624033,0.004170013,0.004908076,0.003477348,0.001217821],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001267333,"about_ca_system_score_gemma":0.001741814,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007340536,"about_ca_topic_score_gemma":0.00859212,"domain_scores_codex":[0.9985083,0.0006370876,0.00009601587,0.0003246147,0.0002781864,0.0001557309],"domain_scores_gemma":[0.9930902,0.00494733,0.0001803643,0.001011052,0.000585457,0.0001855834],"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":[0.0005429452,0.0003782798,0.0009885877,0.0002038554,0.0002107581,0.0001584246,0.0001614371,0.4844719,0.00669523,0.03099335,0.007348977,0.4678463],"study_design_scores_gemma":[0.00001766526,0.00002806318,0.0001244431,0.000006559271,0.00001044409,0.00001531366,0.00001179119,0.9784932,0.001111345,0.01987863,0.0002948594,0.000007834788],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01302718,0.0005910969,0.9837853,0.0002651037,0.00005386607,0.00005989352,0.00007841092,0.001311485,0.0008277288],"genre_scores_gemma":[0.5086374,0.0008523317,0.4770645,0.0005605072,0.0002869557,0.0004027428,0.001253678,0.0008901448,0.01005172],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007340536,"threshold_uncertainty_score":0.01711208,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03693271865177594,"score_gpt":0.2734762916603086,"score_spread":0.2365435730085326,"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."}}