{"id":"W2978190445","doi":"10.1109/qrs.2019.00059","title":"TFCheck : A TensorFlow Library for Detecting Training Issues in Neural Network Programs","year":2019,"lang":"en","type":"article","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Implementation; Machine learning; Artificial intelligence; Training set; Artificial neural network; Code (set theory); Training (meteorology); Process (computing); Focus (optics); Software; Software engineering; Programming language; Set (abstract data type)","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":[0.006916519,0.002357665,0.0009343673,0.002777832,0.001091376,0.002008921,0.003516638,0.001538636,0.008394517],"category_scores_gemma":[0.03344405,0.001504256,0.001791932,0.001197006,0.002545961,0.004755071,0.002628628,0.002596969,0.001747541],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002275935,"about_ca_system_score_gemma":0.004099425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005447205,"about_ca_topic_score_gemma":0.005729823,"domain_scores_codex":[0.9963671,0.0009787563,0.0004698716,0.0007037113,0.001182344,0.0002983011],"domain_scores_gemma":[0.9817224,0.01105989,0.002608639,0.002765683,0.001531251,0.0003121463],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002139364,0.0005270307,0.02834864,0.002368542,0.0004777138,0.001411967,0.001503841,0.2920309,0.04636155,0.05944176,0.06851997,0.4968687],"study_design_scores_gemma":[0.00007859195,0.0001729119,0.001232898,0.000127244,0.00004225288,0.0002281878,0.00005214365,0.9174751,0.04377905,0.02787668,0.008861961,0.00007289358],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0185641,0.0001984198,0.8132144,0.0002748113,0.00009156739,0.0002403963,0.001218123,0.1649,0.001298239],"genre_scores_gemma":[0.2871821,0.0003510394,0.6861417,0.0005181297,0.00008993848,0.0008674308,0.004232688,0.01728222,0.00333471],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008394517,"threshold_uncertainty_score":0.03657854,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01657572878440712,"score_gpt":0.2478104125994643,"score_spread":0.2312346838150572,"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."}}