{"id":"W3206880326","doi":"10.1109/icra48506.2021.9561450","title":"Learning robust driving policies without online exploration","year":2021,"lang":"en","type":"article","venue":"","topic":"Autonomous Vehicle Technology and Safety","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Huawei Technologies (Canada)","funders":"","keywords":"Reinforcement learning; Offline learning; Computer science; Artificial intelligence; Representation (politics); Machine learning; Online learning; Online and offline; Training set; Robot; Scale (ratio)","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":[],"consensus_categories":[],"category_scores_codex":[0.00004584836,0.00007808195,0.00009778882,0.00004282668,0.0000791648,0.0000155364,0.00005356329,0.00009238133,0.000124058],"category_scores_gemma":[0.00003336626,0.00008060308,0.00002420783,0.000138662,0.0000207111,0.0001611347,0.00003307795,0.0002314807,0.00004809792],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002965431,"about_ca_system_score_gemma":0.00001350242,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006023157,"about_ca_topic_score_gemma":0.000190055,"domain_scores_codex":[0.9995771,0.00001281797,0.0001179243,0.00009480177,0.00004698295,0.0001503462],"domain_scores_gemma":[0.9998031,0.00001799643,0.00001152213,0.0001175823,0.00002644353,0.00002336347],"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":[0.00000185596,0.00004167946,0.03850994,0.00002780669,0.00006204143,0.00002147724,0.001094871,0.8850209,0.02614121,0.008655663,0.0004287052,0.03999385],"study_design_scores_gemma":[0.0004003172,0.00004658335,0.0264611,0.00004782438,0.00002433425,0.00005020383,0.005301389,0.8514354,0.09678057,0.001051131,0.01790716,0.0004939782],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.866243,0.0001164995,0.1174632,0.0007065094,0.0001058585,0.00003528446,7.987267e-7,0.002082826,0.013246],"genre_scores_gemma":[0.9913573,0.0001149455,0.006262571,0.00004771939,0.00005251445,0.000001606105,0.0000220386,0.00001866963,0.00212269],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1251142,"threshold_uncertainty_score":0.3286898,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01796128547711093,"score_gpt":0.2250142006271975,"score_spread":0.2070529151500866,"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."}}