{"id":"W3132089310","doi":"10.1109/iros45743.2020.9341019","title":"Learning Domain Randomization Distributions for Training Robust Locomotion Policies","year":2020,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Artificial intelligence; Context (archaeology); Machine learning; Task (project management); Generalization; Extrapolation; Domain (mathematical analysis); Interpolation (computer graphics); Range (aeronautics); Mathematics; Statistics; 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":[],"consensus_categories":[],"category_scores_codex":[0.0002984759,0.0001075691,0.0001530405,0.00004701697,0.0003276796,0.0002251477,0.000335899,0.0000474113,0.00002194214],"category_scores_gemma":[0.0004733552,0.000102221,0.00008087132,0.0003636348,0.00002843658,0.0004050362,0.00009678172,0.0001270815,0.00002261068],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003947231,"about_ca_system_score_gemma":0.00004407368,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004194711,"about_ca_topic_score_gemma":7.779673e-7,"domain_scores_codex":[0.9990218,0.00007364684,0.0002477261,0.0002271247,0.0001726192,0.0002571412],"domain_scores_gemma":[0.9993808,0.0001862762,0.0001123289,0.0001338953,0.00008943059,0.00009723442],"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":[0.0000118076,0.000004170461,0.00009793873,0.0000100758,0.000009814865,2.827917e-7,0.002663416,0.8902642,0.0002569434,0.1042798,0.0005021542,0.001899379],"study_design_scores_gemma":[0.001727668,0.0001537067,0.0001057278,0.00000884429,0.000006803425,0.000001886362,0.000416296,0.988465,0.0002862322,0.0005634004,0.0081328,0.0001316867],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007372155,0.000005502161,0.9887339,0.007565741,0.0001033672,0.0003478607,0.000001264775,0.0004249464,0.002080189],"genre_scores_gemma":[0.7170074,0.000005109162,0.2817869,0.0004903669,0.0001447797,0.00003495941,0.00006855028,0.0000115538,0.0004503297],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7162702,"threshold_uncertainty_score":0.4168453,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05326927639304079,"score_gpt":0.25799983112925,"score_spread":0.2047305547362092,"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."}}