{"id":"W4386875701","doi":"10.48550/arxiv.2309.09476","title":"Mechanic Maker 2.0: Reinforcement Learning for Evaluating Generated Rules","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Artificial Intelligence in Games","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alberta Machine Intelligence Institute; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"USable; Reinforcement learning; Computer science; Artificial intelligence; Meaning (existential); Machine learning; Reinforcement; Baseline (sea); Psychology; World Wide Web; Social psychology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009398928,0.0003483382,0.0003433996,0.0003176843,0.0003830702,0.0002444388,0.001775307,0.0002908732,0.00004794243],"category_scores_gemma":[0.0003378539,0.0004144158,0.0002663289,0.0005227614,0.00006082192,0.0002932904,0.002405721,0.0005644123,0.0004813339],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003033817,"about_ca_system_score_gemma":0.0002680515,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001777731,"about_ca_topic_score_gemma":0.0000446711,"domain_scores_codex":[0.9974335,0.0001827802,0.0003773722,0.001275748,0.0001811468,0.0005494356],"domain_scores_gemma":[0.9977368,0.0003261185,0.0004042268,0.0009678944,0.0004328507,0.0001321578],"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":[0.00001669259,0.00001358178,0.00009635121,0.00004899773,0.0000768954,0.00003912005,0.0002990834,0.8776464,0.0003477834,0.1188124,0.0002031066,0.002399536],"study_design_scores_gemma":[0.0001162433,0.0001141503,0.00002627947,0.00009971489,0.00005299522,9.698904e-7,0.0001566482,0.8978752,0.002764434,0.09809977,0.0002932565,0.0004002778],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1004319,0.00002877659,0.8966489,0.0001181934,0.001115753,0.0006293746,0.000004065457,0.0006807426,0.0003422815],"genre_scores_gemma":[0.9779825,0.00009176969,0.01158414,0.00007536048,0.0001344756,0.00001244877,0.00005247121,0.00004413186,0.01002268],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8850648,"threshold_uncertainty_score":0.9998308,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.265515225508008,"score_gpt":0.2751630985045367,"score_spread":0.009647872996528695,"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."}}