{"id":"W4392542471","doi":"10.1145/3626252.3630759","title":"A Fast and Accurate Machine Learning Autograder for the Breakout Assignment","year":2024,"lang":"en","type":"article","venue":"","topic":"Teaching and Learning Programming","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research; Universitas Brawijaya; National Science Foundation","keywords":"Breakout; Grading (engineering); Computer science; Software deployment; Unit testing; Artificial intelligence; Reinforcement learning; Paddle; Machine learning; Multimedia; Software engineering; Operating system; Engineering; Software","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006486803,0.00009976952,0.0000743765,0.00004689852,0.0003602385,0.0009958819,0.0002831631,0.00002846255,0.00001263647],"category_scores_gemma":[0.00005036739,0.00005731856,0.00005222964,0.0001420463,0.00002847888,0.0001799823,0.0001482415,0.0003040607,0.00001395182],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001429456,"about_ca_system_score_gemma":0.00002176802,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001116876,"about_ca_topic_score_gemma":0.000008831311,"domain_scores_codex":[0.9992017,0.00006495446,0.0001026723,0.0002873101,0.0001242598,0.0002190739],"domain_scores_gemma":[0.9992412,0.000510523,0.00002204853,0.0001680971,0.0000110788,0.00004709377],"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":[0.000001231284,0.000006880181,0.0001937896,0.00001876063,0.00002851771,0.000002887623,0.002449437,0.001028225,0.00003122801,0.02420437,0.0003545991,0.97168],"study_design_scores_gemma":[0.00007078119,0.0000535177,0.0002871367,0.0000185103,0.000008032014,0.00002048434,0.00009592871,0.5723193,0.00001467666,0.0001170887,0.4269245,0.0000700083],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.000334043,0.001945126,0.9883024,0.007919971,0.0002697961,0.0001523374,4.92319e-7,0.0007119258,0.0003639564],"genre_scores_gemma":[0.9319059,0.00002451445,0.05799126,0.0002057239,0.00008541719,0.00004853673,0.00000165732,0.00001383801,0.00972315],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9716101,"threshold_uncertainty_score":0.9603311,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02167138927170486,"score_gpt":0.2766019710978808,"score_spread":0.2549305818261759,"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."}}