{"id":"W2737821837","doi":"10.1109/icra.2017.7989686","title":"Adapting learned robotics behaviours through policy adjustment","year":2017,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Robot; Computer science; Inverted pendulum; Task (project management); Artificial intelligence; Robotics; Action (physics); State space; Physical system; Mobile robot; Control theory (sociology); Control (management); Engineering; Mathematics","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":[0.0002427594,0.0001927756,0.0001850221,0.00007002983,0.0007714516,0.0007831174,0.002365692,0.00008445605,0.00003281471],"category_scores_gemma":[0.0002707269,0.0001757278,0.00008969547,0.0001063632,0.0001010076,0.001191888,0.00113597,0.0002314161,0.0002018597],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001168499,"about_ca_system_score_gemma":0.0001441719,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007722023,"about_ca_topic_score_gemma":0.00001664241,"domain_scores_codex":[0.9983184,0.0000398615,0.0002770701,0.0003905709,0.0004561705,0.0005178901],"domain_scores_gemma":[0.997529,0.00005212981,0.0003080918,0.001909343,0.00008832567,0.000113131],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001557991,0.00002500436,0.002383928,0.000004743812,0.00001764882,0.00001575309,0.0004044371,0.4438893,0.00004357464,0.5409667,0.0007982485,0.01144906],"study_design_scores_gemma":[0.001172303,0.0003621241,0.04258855,0.00007282089,0.00003487618,0.00004797214,0.0001361036,0.942005,0.00105321,0.005502697,0.006191665,0.0008326378],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0004798458,0.00002663324,0.917501,0.007152887,0.0005847751,0.0001511897,2.857028e-7,0.0002883641,0.07381506],"genre_scores_gemma":[0.5595465,0.00005529893,0.421158,0.0009822764,0.0003591623,0.00000759227,0.000002761867,0.0000257327,0.01786257],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5590667,"threshold_uncertainty_score":0.7551619,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07174920135323527,"score_gpt":0.3402462901202425,"score_spread":0.2684970887670072,"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."}}