{"id":"W7124296281","doi":"10.65109/fkdg1872","title":"Minimax Exploiter: A Data Efficient Approach for Competitive Self-Play","year":2024,"lang":"","type":"article","venue":"","topic":"Artificial Intelligence in Games","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ubisoft (Canada); McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Minimax; Key (lock); Reinforcement learning; Core (optical fiber); Stability (learning theory); Simple (philosophy); Component (thermodynamics); Game theory","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":["metaepi_narrow","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001877468,0.000666261,0.0006036515,0.0003866528,0.0004343779,0.002800191,0.005812914,0.0002530139,0.0004987252],"category_scores_gemma":[0.0002161399,0.0005965166,0.000299062,0.001348112,0.0004036891,0.001301246,0.003561175,0.0004513827,0.00155979],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002175895,"about_ca_system_score_gemma":0.0006812846,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004300801,"about_ca_topic_score_gemma":0.00001426814,"domain_scores_codex":[0.9933067,0.0001989523,0.001172167,0.003191633,0.0008796123,0.001250923],"domain_scores_gemma":[0.9942957,0.001457711,0.0001499856,0.003371039,0.0003536685,0.0003719367],"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":[0.00007090274,0.002059598,0.00003350368,0.0009160797,0.0006307957,0.0001125198,0.03695882,0.005149149,0.0004336942,0.6511857,0.03049552,0.2719537],"study_design_scores_gemma":[0.00009476469,0.0003072421,0.000005458839,0.000206917,0.0001144771,0.00004868903,0.002577659,0.8857225,0.004804777,0.0009358839,0.1045016,0.0006800157],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007590996,0.003263496,0.9553727,0.001559239,0.004050538,0.001964654,0.0004009005,0.0007281702,0.03190121],"genre_scores_gemma":[0.5008324,0.0001592808,0.4932428,0.0004182908,0.0009679249,0.0002024323,0.0001516737,0.0000831639,0.003942083],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8805734,"threshold_uncertainty_score":0.9996486,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09681424090509325,"score_gpt":0.3333619045821295,"score_spread":0.2365476636770363,"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."}}