{"id":"W4389482477","doi":"10.54254/2753-8818/19/20230542","title":"An evaluation of reinforcement learning performance in the iterated prisoner’s dilemma","year":2023,"lang":"en","type":"article","venue":"Theoretical and Natural Science","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"University of Cambridge","keywords":"Reinforcement learning; Dilemma; Iterated function; Prisoner's dilemma; Reinforcement; Computer science; Artificial neural network; Artificial intelligence; Psychology; Social psychology; 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":[0.005875229,0.00008032699,0.00008648589,0.000158033,0.0002240398,0.0001579912,0.000916195,0.00002581258,0.00001369282],"category_scores_gemma":[0.000408822,0.00004772685,0.00001372127,0.001843088,0.0007572473,0.000718865,0.0001971275,0.0002316549,0.00001065998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003192434,"about_ca_system_score_gemma":0.00008680142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004817482,"about_ca_topic_score_gemma":3.198485e-7,"domain_scores_codex":[0.9979134,0.0001956502,0.0002050929,0.0002423571,0.00114636,0.0002971857],"domain_scores_gemma":[0.9992829,0.0001468399,0.0000668538,0.0002683876,0.000185446,0.00004958871],"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":[0.000008078398,0.000007329019,0.001977736,0.000007717783,0.000001043192,9.109244e-7,0.001871933,0.4512439,0.00248235,0.5258546,0.000002632169,0.01654175],"study_design_scores_gemma":[0.0001734938,0.0001870296,0.04896823,0.00002233627,0.000003265027,0.000003469582,0.00007608065,0.9470074,0.001369421,0.002086489,0.00003137153,0.00007140871],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9829789,0.00001890026,0.01462192,0.0004043957,0.0001063317,0.0002002079,4.868188e-8,0.00005431203,0.001614933],"genre_scores_gemma":[0.9994277,0.00002141223,0.0004317785,0.00007236525,0.00001208097,0.000007727088,0.000002292667,0.000002021666,0.0000226331],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5237681,"threshold_uncertainty_score":0.2790109,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01822272713279608,"score_gpt":0.2948606301005942,"score_spread":0.2766379029677981,"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."}}