{"id":"W4293741188","doi":"10.3389/frobt.2022.854212","title":"Model-free reinforcement learning for robust locomotion using demonstrations from trajectory optimization","year":2022,"lang":"en","type":"article","venue":"Frontiers in Robotics and AI","topic":"Robotic Locomotion and Control","field":"Engineering","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"European Commission; York University; National Science Foundation","keywords":"Computer science; Reinforcement learning; Bounding overwatch; Robustness (evolution); Robot; Trajectory optimization; Trajectory; Task (project management); Artificial intelligence","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001130221,0.0001102714,0.0001510046,0.0001264697,0.0002375071,0.00004286604,0.00009296698,0.00004574898,0.00002258199],"category_scores_gemma":[0.00001322709,0.0001355977,0.00004069736,0.0001124987,0.00001728269,0.0001212366,0.00003428684,0.0001996508,1.123649e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001617248,"about_ca_system_score_gemma":0.00002634798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000121656,"about_ca_topic_score_gemma":0.000006452093,"domain_scores_codex":[0.99931,0.00002834368,0.0002298476,0.0001440998,0.0001161234,0.0001716192],"domain_scores_gemma":[0.999762,0.00001558869,0.00003734929,0.0001185149,0.0000216951,0.00004483894],"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":[0.000008344681,0.00001671284,0.0005775402,0.00001256406,0.00002406955,5.875179e-7,0.0001680674,0.9967045,0.0000575473,0.0003207744,0.001156443,0.0009528816],"study_design_scores_gemma":[0.000954025,0.00003712669,0.00004269179,0.00001149024,0.00003284588,9.37628e-7,0.0002771145,0.9976419,0.000008128333,0.0007806555,0.00006690076,0.0001462145],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0008725609,0.000240405,0.9976465,0.0001875975,0.000492972,0.0003182467,0.000009199547,0.00007340174,0.0001591482],"genre_scores_gemma":[0.5305814,0.00005841147,0.468925,0.00007480181,0.00004981223,0.00005962088,0.0001099744,0.00002927242,0.0001117128],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5297088,"threshold_uncertainty_score":0.5529515,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01804940780921152,"score_gpt":0.207816876286318,"score_spread":0.1897674684771065,"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."}}