{"id":"W3004493192","doi":"10.1016/j.jss.2020.110542","title":"On testing machine learning programs","year":2020,"lang":"en","type":"article","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":198,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Transformative learning; Domain (mathematical analysis); Software testing; Software engineering; Test strategy; Field (mathematics); Software; Witness; Data science; Artificial intelligence; Machine learning; Psychology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006225286,0.0003966389,0.0003570227,0.0002407508,0.0003784891,0.0004866751,0.001447818,0.0001923467,0.000008407723],"category_scores_gemma":[0.003084274,0.0003855487,0.00013793,0.001408991,0.00006685615,0.0003735695,0.0005781389,0.000893072,0.00005395525],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001680288,"about_ca_system_score_gemma":0.0001408759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002345767,"about_ca_topic_score_gemma":0.0000608046,"domain_scores_codex":[0.9973204,0.0001822147,0.0004258977,0.0007830232,0.0004956172,0.0007928479],"domain_scores_gemma":[0.9977366,0.0005628861,0.0002461784,0.0008289374,0.0001443587,0.000481016],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005925542,0.0003617396,0.1346812,0.00008152946,0.00004391948,0.0003630303,0.000920746,0.004382005,0.002162282,0.07154898,0.007626771,0.7777686],"study_design_scores_gemma":[0.0002777474,0.00102416,0.003028185,0.000115515,0.000009612141,0.0001086604,0.000004947976,0.9702724,0.002385322,0.02046677,0.001792184,0.0005145049],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01382433,0.0003209152,0.9335827,0.005797466,0.00006663262,0.0004934,0.000003068971,0.04488453,0.00102702],"genre_scores_gemma":[0.5403733,0.000008199279,0.4547417,0.004517992,0.00008211557,0.000161815,0.000007172563,0.00004429712,0.00006343784],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9658904,"threshold_uncertainty_score":0.9998596,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02772335947957308,"score_gpt":0.238025687806026,"score_spread":0.2103023283264529,"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."}}