{"id":"W4405602545","doi":"10.1109/icsme58944.2024.00019","title":"Toward Debugging Deep Reinforcement Learning Programs with RLExplorer","year":2024,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Debugging; Reinforcement learning; Computer science; Algorithmic program debugging; Programming language; Artificial intelligence; Human–computer interaction; Software engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0002925438,0.0002246251,0.0001563693,0.0001695818,0.0001493841,0.001113686,0.0006586008,0.00005353013,0.0001151545],"category_scores_gemma":[0.00002340572,0.0001618472,0.00007020561,0.0006487789,0.00004716589,0.0009369743,0.0003228016,0.000383478,0.0004727395],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001098484,"about_ca_system_score_gemma":0.00009867687,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001907135,"about_ca_topic_score_gemma":0.000002405924,"domain_scores_codex":[0.9980821,0.0000367481,0.000275659,0.0004810095,0.0005972069,0.0005272812],"domain_scores_gemma":[0.9992643,0.00006638492,0.00005382002,0.0004149719,0.00007115792,0.0001293404],"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":[0.000003533825,0.000006523719,0.000337963,0.00006555524,0.00006333589,0.00007998675,0.002121544,0.8897516,0.00002386924,0.05698769,0.0001793051,0.05037914],"study_design_scores_gemma":[0.0001481903,0.0003854856,0.00003133493,0.0001371283,0.00001429491,0.00004355408,0.0001552429,0.9676232,0.0002661815,0.000129923,0.03079687,0.0002686491],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0002478657,0.0001398065,0.9575876,0.000819607,0.000293777,0.0002578915,7.199132e-9,0.002239997,0.03841347],"genre_scores_gemma":[0.9134021,0.00002085454,0.07743005,0.0002656351,0.00007361204,0.00005310401,0.000007031175,0.00003055503,0.008717066],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9131542,"threshold_uncertainty_score":0.9999232,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03009305622501012,"score_gpt":0.2465751540760077,"score_spread":0.2164820978509975,"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."}}