{"id":"W137325057","doi":"10.1609/aaai.v26i1.8319","title":"Approximate Policy Iteration with Linear Action Models","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates - Technology Futures","keywords":"Benchmark (surveying); Computer science; Action (physics); Mathematical optimization; Linear model; Machine learning; 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":[0.003064049,0.001276325,0.002417265,0.0007341575,0.0004350848,0.001446114,0.00231246,0.002218971,0.002829545],"category_scores_gemma":[0.01262128,0.001062221,0.0007671948,0.0009814098,0.001749696,0.002452303,0.001783129,0.003063324,0.0007347279],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001608278,"about_ca_system_score_gemma":0.002860594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007369473,"about_ca_topic_score_gemma":0.006346474,"domain_scores_codex":[0.9980427,0.0008632961,0.0001032745,0.0003878913,0.0004190932,0.0001836586],"domain_scores_gemma":[0.9928606,0.005646829,0.0004441513,0.0004236865,0.0004421734,0.0001824218],"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":[0.0001123917,0.00005926314,0.0003918467,0.00006920936,0.00003638746,0.00003419335,0.0000401631,0.9667376,0.0003245191,0.007117634,0.000464003,0.02461289],"study_design_scores_gemma":[0.000008104369,0.00001860051,0.00001758057,0.00000321405,0.000002101513,0.000004716902,0.000002794075,0.9954425,0.000154559,0.004213067,0.0001299907,0.000002743584],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01219075,0.0003272342,0.9851608,0.0002207296,0.00003604243,0.00005170739,0.00005122731,0.0006756426,0.001285943],"genre_scores_gemma":[0.6353007,0.0002787504,0.3582512,0.0003608423,0.00007999519,0.0004598206,0.0003948073,0.000220418,0.004653445],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007369473,"threshold_uncertainty_score":0.01620442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.12002286483175,"score_gpt":0.3180867896487678,"score_spread":0.1980639248170178,"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."}}