{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002370014,0.0005921039,0.0004750386,0.0005922841,0.0002413041,0.0005603937,0.001163471,0.0007726821,0.001221154],"category_scores_gemma":[0.01457914,0.0002163261,0.0004180023,0.0002268486,0.001408637,0.001006236,0.0007552023,0.0008521659,0.0001077989],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001413574,"about_ca_system_score_gemma":0.001088443,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003163506,"about_ca_topic_score_gemma":0.002192027,"domain_scores_codex":[0.9984494,0.0005661349,0.00008602033,0.0002903605,0.0003564342,0.0002514831],"domain_scores_gemma":[0.9903311,0.006956156,0.0007513129,0.000802758,0.0008590407,0.0002996635],"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":[0.0004000986,0.0001974237,0.009889469,0.0001317915,0.00006540561,0.0003616132,0.0001332344,0.9067916,0.01082827,0.01166437,0.0006639678,0.05887271],"study_design_scores_gemma":[0.0000117496,0.00007464879,0.0003271823,0.000008206736,0.000007302759,0.00002901671,0.00000964435,0.9918157,0.003466993,0.004135957,0.0001090825,0.000004472435],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6330793,0.000219276,0.3612432,0.0004200529,0.0000589254,0.0000955521,0.0001178609,0.001981849,0.002783785],"genre_scores_gemma":[0.9796146,0.00001773202,0.01981809,0.00005756763,0.000003616468,0.00002882592,0.00004433971,0.00003253229,0.000382821],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.003163506,"threshold_uncertainty_score":0.01253396,"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."}}