{"id":"W4378418585","doi":"10.1109/icst57152.2023.00026","title":"Mutation Testing of Deep Reinforcement Learning Based on Real Faults","year":2023,"lang":"en","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Science and Engineering Research Council; Bombardier","keywords":"Reinforcement learning; Mutation; Computer science; Heuristic; Set (abstract data type); Artificial intelligence; Task (project management); Machine learning; Adaptive mutation; Order (exchange); Genetic algorithm; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0004274118,0.00007017656,0.0000792721,0.0001773182,0.0000825234,0.00002413553,0.0002325716,0.00002832863,0.000003961265],"category_scores_gemma":[0.001197516,0.00006303541,0.00002412248,0.0007240097,0.00001301528,0.00008541766,0.00007517685,0.00007630958,0.0000282164],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002291464,"about_ca_system_score_gemma":0.00002814721,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001921329,"about_ca_topic_score_gemma":0.000001564526,"domain_scores_codex":[0.9992093,0.00003471045,0.000173068,0.0001820603,0.0002488142,0.0001520183],"domain_scores_gemma":[0.9986562,0.0008857591,0.00009258034,0.0002418067,0.00009271154,0.00003096022],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006401848,0.00002342665,0.0136753,0.00003799743,0.000004152577,0.00002861156,0.0005187435,0.7991214,0.0003490259,0.002610154,0.001458274,0.1821666],"study_design_scores_gemma":[0.0001021663,0.0002564442,0.005190562,0.00005430623,0.00000121028,0.000001585282,0.000005798815,0.989194,0.00283407,0.002271682,0.00001549756,0.0000726373],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01106474,7.761427e-7,0.9616395,0.0001050994,0.00005119669,0.00007585256,4.86801e-8,0.01433943,0.01272336],"genre_scores_gemma":[0.8145471,3.780046e-7,0.185207,0.00007752103,0.00001056693,0.000008833839,0.000003778062,0.00000507255,0.000139814],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8034823,"threshold_uncertainty_score":0.257051,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03580877730820999,"score_gpt":0.2859746181044491,"score_spread":0.2501658407962391,"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."}}