{"id":"W2903495827","doi":"10.1016/j.jss.2020.110542","title":"On testing machine learning programs","year":2020,"lang":"en","type":"preprint","venue":"Journal of Systems and Software","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Transformative learning; Domain (mathematical analysis); Witness; Software engineering; Field (mathematics); Reliability (semiconductor); Software testing; Artificial intelligence; Software; Data science; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007477098,0.003147743,0.002845463,0.005191264,0.001364154,0.002195015,0.005030093,0.003794504,0.009724172],"category_scores_gemma":[0.08197311,0.001592562,0.002086336,0.00488133,0.00654039,0.01256323,0.006170699,0.005180425,0.000886591],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00256644,"about_ca_system_score_gemma":0.001708046,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005622361,"about_ca_topic_score_gemma":0.006447045,"domain_scores_codex":[0.9843576,0.007791942,0.0009528748,0.001880817,0.004024982,0.0009917675],"domain_scores_gemma":[0.8295326,0.1487401,0.001753405,0.01440582,0.00442986,0.001138251],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001454409,0.0006768494,0.009906262,0.0007581528,0.0003618979,0.0007321168,0.0005357806,0.2025197,0.004993046,0.2158561,0.02385618,0.5383495],"study_design_scores_gemma":[0.0001096284,0.0001727905,0.002048625,0.000177276,0.0001520258,0.0002531628,0.0001132516,0.4859421,0.003438666,0.5036209,0.003937749,0.0000338435],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06693225,0.004021714,0.8977884,0.005043826,0.0007252413,0.0001866457,0.0003938551,0.002329713,0.0225784],"genre_scores_gemma":[0.6919882,0.002165932,0.2865965,0.001613507,0.001410007,0.0003129737,0.001666881,0.001234238,0.01301178],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009724172,"threshold_uncertainty_score":0.03954315,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06147200893269984,"score_gpt":0.2724679788593244,"score_spread":0.2109959699266246,"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."}}