{"id":"W4285102525","doi":"10.1109/icra46639.2022.9811652","title":"Exploiting Abstract Symmetries in Reinforcement Learning for Complex Environments","year":2022,"lang":"en","type":"article","venue":"2022 International Conference on Robotics and Automation (ICRA)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia, Okanagan Campus; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Computer science; Heuristics; Abstraction; Exploit; Sample (material); Artificial intelligence; Sample complexity; State space; Inefficiency; Space (punctuation); Field (mathematics); State (computer science); Algorithm; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001963367,0.0006032199,0.0008847088,0.000380572,0.0003495047,0.0009272749,0.0009026999,0.0005960043,0.001090993],"category_scores_gemma":[0.00583332,0.000347657,0.0007340816,0.0003063836,0.002026792,0.002340419,0.002701586,0.001959893,0.0001937934],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006968204,"about_ca_system_score_gemma":0.000783419,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009296409,"about_ca_topic_score_gemma":0.0009558754,"domain_scores_codex":[0.9992852,0.000323406,0.0000509768,0.0001050216,0.0001750144,0.00006048774],"domain_scores_gemma":[0.9976751,0.001247674,0.0003118286,0.0004579904,0.0001275744,0.0001797239],"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":[0.0001657047,0.00009774658,0.001656517,0.0001532553,0.00006220218,0.0001645782,0.0003027444,0.6756751,0.006648864,0.2135358,0.0006581088,0.1008794],"study_design_scores_gemma":[0.00003748303,0.0001001225,0.0001918554,0.00001439356,0.000009784394,0.00004752953,0.0000207762,0.8545346,0.001682996,0.1422826,0.001061347,0.00001643437],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01545313,0.00008919811,0.983349,0.0001198798,0.00001443309,0.00003529188,0.00001139141,0.0001608601,0.0007667971],"genre_scores_gemma":[0.6857373,0.0002512989,0.3125244,0.0001097349,0.00003888388,0.000172511,0.00005136962,0.0000621006,0.001052332],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001963367,"threshold_uncertainty_score":0.01038337,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06210813645581727,"score_gpt":0.2900027725995281,"score_spread":0.2278946361437108,"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."}}