{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002702127,0.001318034,0.001671932,0.001037347,0.0007509263,0.001353083,0.003736208,0.001661876,0.004817768],"category_scores_gemma":[0.00687336,0.0007260221,0.0007757029,0.0006696038,0.001479341,0.001948119,0.0033008,0.002374825,0.001026267],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009543679,"about_ca_system_score_gemma":0.002020105,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003025145,"about_ca_topic_score_gemma":0.0036504,"domain_scores_codex":[0.9988816,0.000413968,0.00005456043,0.0001762329,0.0003608349,0.0001128682],"domain_scores_gemma":[0.9975286,0.00149912,0.0002186385,0.000298032,0.0003084691,0.0001470856],"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":[0.0003277737,0.0001982257,0.001455048,0.0001763466,0.0001273521,0.0001721796,0.0001369967,0.8221288,0.003367969,0.04375595,0.005057308,0.1230961],"study_design_scores_gemma":[0.00001638511,0.00003536054,0.00004226734,0.00000592376,0.000003737373,0.0000171603,0.000007037807,0.9920118,0.0003833874,0.006740544,0.0007317921,0.000004572624],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01149963,0.0002672827,0.9837474,0.0003042723,0.0000451083,0.0001143158,0.00007212867,0.001192602,0.002757305],"genre_scores_gemma":[0.4356254,0.0002785493,0.5550255,0.0004222859,0.0001043953,0.0006686662,0.0003310131,0.000643035,0.006901206],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004817768,"threshold_uncertainty_score":0.01611698,"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."}}