{"id":"W2946807670","doi":"10.1109/cig.2019.8848125","title":"Learning Policies from Human Data for Skat","year":2019,"lang":"en","type":"preprint","venue":"2019 IEEE Conference on Games (CoG)","topic":"Sports Analytics and Performance","field":"Economics, Econometrics and Finance","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Counterfactual thinking; Reinforcement learning; Perfect information; Bidding; Regret; Construct (python library); Counterfactual conditional; Artificial intelligence; State (computer science); Imitation; Value (mathematics); Action (physics); Bridge (graph theory); Task (project management); Data science; Machine learning; Microeconomics; Economics","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0005914683,0.0005137186,0.001144953,0.0003395716,0.0001743179,0.0005202747,0.001827764,0.0004941262,0.001483505],"category_scores_gemma":[0.00008972688,0.0005654345,0.0002472283,0.00008513073,0.0001042323,0.0002208487,0.0007245648,0.0009220922,0.001620968],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007601805,"about_ca_system_score_gemma":0.0001395648,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003778021,"about_ca_topic_score_gemma":0.0002262666,"domain_scores_codex":[0.9968817,0.00001527254,0.0009709333,0.001471763,0.0001009103,0.0005593848],"domain_scores_gemma":[0.9959399,0.0001066602,0.001176982,0.002529139,0.0001197714,0.0001275219],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003149114,0.0008007578,0.1790877,0.001423681,0.002670733,0.00001605906,0.004876401,0.04688935,0.0002802465,0.4683461,0.2741239,0.02117013],"study_design_scores_gemma":[0.001401004,0.0004628423,0.04276609,0.0007077313,0.0001633465,0.000001054045,0.000250925,0.3430129,0.0001034048,0.05942035,0.549543,0.002167392],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8228692,0.005688183,0.01553193,0.002758784,0.01100608,0.003365677,0.03205455,0.0003491787,0.1063764],"genre_scores_gemma":[0.9647405,0.001697414,0.0002949713,0.0003987387,0.0009381811,0.00006763153,0.004191417,0.00009083397,0.02758036],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4089257,"threshold_uncertainty_score":0.9996797,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1733676294669931,"score_gpt":0.314959953163113,"score_spread":0.1415923236961198,"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."}}