{"id":"W4386185028","doi":"10.48550/arxiv.2308.12438","title":"Deploying Deep Reinforcement Learning Systems: A Taxonomy of Challenges","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Open Source Software Innovations","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Software deployment; Reinforcement learning; Computer science; Enthusiasm; Artificial intelligence; Autonomy; Data science; Software engineering; Political science; Psychology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005461981,0.0002880594,0.0004260734,0.0005717367,0.0001806813,0.0001206522,0.002005265,0.0002384616,0.000008003438],"category_scores_gemma":[0.0001249686,0.0003603609,0.0001654015,0.0009317258,0.0000806544,0.0003879413,0.003347869,0.0006586512,0.000130378],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002548617,"about_ca_system_score_gemma":0.0002014832,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00029597,"about_ca_topic_score_gemma":0.00003103562,"domain_scores_codex":[0.997977,0.000163587,0.0004070773,0.0009319115,0.0001611395,0.0003592946],"domain_scores_gemma":[0.9973494,0.0002490925,0.0006532833,0.001275842,0.0003733625,0.00009897305],"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":[0.000003737617,0.00001903106,0.001649515,0.0002377756,0.0001146928,0.00006954286,0.0004863087,0.8218523,0.000005059833,0.1749947,0.00003925796,0.0005280564],"study_design_scores_gemma":[0.0003423949,0.00007208154,0.0006352013,0.0005082256,0.000058294,0.000004259742,0.001271555,0.9908404,0.00004360835,0.003502836,0.002202157,0.0005189955],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01850299,0.000200079,0.9766339,0.00009184224,0.0006664386,0.0006183663,0.00000163271,0.0007432702,0.002541467],"genre_scores_gemma":[0.9952738,0.0003348194,0.00256873,0.00001194903,0.00005446468,0.00001182089,0.00001049896,0.0000313494,0.001702551],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9767708,"threshold_uncertainty_score":0.9998848,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1837989261232988,"score_gpt":0.2076928440857742,"score_spread":0.02389391796247539,"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."}}