{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005030281,0.0001590178,0.0001602643,0.0003376458,0.0003855582,0.0002824231,0.0005697622,0.00003246815,0.0003631887],"category_scores_gemma":[0.0001426652,0.0001836574,0.00004249708,0.0001773587,0.0000325257,0.0003462773,0.0004474547,0.0003077523,0.00001419299],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002360134,"about_ca_system_score_gemma":0.0000519887,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001123444,"about_ca_topic_score_gemma":0.000001323807,"domain_scores_codex":[0.9982243,0.00005677925,0.0004461925,0.000348205,0.000682295,0.0002421819],"domain_scores_gemma":[0.9992145,0.0001695623,0.0003145973,0.0001809132,0.00006828429,0.00005215112],"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":[0.000008866876,0.00002829985,0.0004564863,0.000009081074,0.00001700858,0.000002852527,0.0002821871,0.8038387,0.0009479963,0.1910961,0.0001305265,0.003181916],"study_design_scores_gemma":[0.0005991872,0.000235787,0.005521636,0.00002492439,0.000003467357,0.000004168478,0.0004414555,0.9881295,0.0001160905,0.001243896,0.003490121,0.0001897445],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003571885,0.00001027362,0.9862233,0.003319288,0.0005125958,0.0003694795,0.000004720918,0.00009171529,0.00589677],"genre_scores_gemma":[0.9870206,0.00004339735,0.01029495,0.0004672955,0.00003292205,0.0001121735,0.0001622834,0.0000127588,0.00185362],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9834487,"threshold_uncertainty_score":0.748933,"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."}}