{"id":"W4281553483","doi":"10.24963/ijcai.2022/484","title":"Approximate Exploitability: Learning a Best Response","year":2022,"lang":"en","type":"article","venue":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","topic":"Artificial Intelligence in Games","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Adversarial system; Computer science; Reinforcement learning; Scalability; Robustness (evolution); Artificial intelligence; Machine learning; Variety (cybernetics); Computation; Deep neural networks; Artificial neural network; Theoretical computer science; Algorithm","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":[0.004515212,0.001274276,0.001566482,0.0008262006,0.0007199606,0.001479419,0.002309582,0.002058557,0.003630368],"category_scores_gemma":[0.02876958,0.0007097049,0.0008323626,0.0006524266,0.002628061,0.003430409,0.00291324,0.00327047,0.0006234412],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001979223,"about_ca_system_score_gemma":0.001806253,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002990262,"about_ca_topic_score_gemma":0.002955087,"domain_scores_codex":[0.9965993,0.001588193,0.0001438449,0.0007115098,0.000551183,0.0004060226],"domain_scores_gemma":[0.9870856,0.009275021,0.001079712,0.001388881,0.0006821349,0.0004885893],"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":[0.0003351654,0.0001347643,0.001966277,0.00009348882,0.00008739592,0.0001053474,0.0001844516,0.9172845,0.001585653,0.03992362,0.002419276,0.03588011],"study_design_scores_gemma":[0.00002124188,0.00006111879,0.0001105185,0.00001242817,0.000008558197,0.00002288853,0.00002305233,0.9644516,0.0004200281,0.03451675,0.0003443053,0.00000742329],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07635602,0.0003334707,0.9152833,0.001139089,0.00006665567,0.0001232949,0.0001345306,0.0009003461,0.005663247],"genre_scores_gemma":[0.9102951,0.0001504038,0.08596006,0.0003721259,0.0000454781,0.0002171508,0.0001673557,0.0001580369,0.002634248],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004515212,"threshold_uncertainty_score":0.02387905,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08809639599173266,"score_gpt":0.2973602486952115,"score_spread":0.2092638527034789,"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."}}