{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005438473,0.002412608,0.001689657,0.003910051,0.001144177,0.001782613,0.004615527,0.002585734,0.008551338],"category_scores_gemma":[0.06106046,0.001278825,0.00204793,0.003476035,0.003617962,0.00854791,0.004396002,0.003399719,0.001070421],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002187337,"about_ca_system_score_gemma":0.001854638,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008490797,"about_ca_topic_score_gemma":0.01206508,"domain_scores_codex":[0.9829312,0.007889721,0.0009614869,0.002269737,0.004868865,0.001078994],"domain_scores_gemma":[0.8986333,0.08192474,0.001594327,0.01248297,0.00449524,0.0008694676],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001033769,0.0006561562,0.01037528,0.0006831214,0.0003051084,0.0006826919,0.0004967625,0.1292358,0.009342832,0.121542,0.02142987,0.7042165],"study_design_scores_gemma":[0.0001266216,0.0002885397,0.002589931,0.0002325501,0.0001736887,0.0004388753,0.0001534572,0.6812806,0.01229817,0.2944688,0.007908997,0.00003979039],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07285555,0.002247791,0.8975637,0.002961765,0.0003947412,0.0001768369,0.0003821117,0.006710056,0.01670742],"genre_scores_gemma":[0.607393,0.001010585,0.3744785,0.001334959,0.000487614,0.0002521727,0.001944003,0.001838176,0.01126105],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008551338,"threshold_uncertainty_score":0.02876174,"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."}}