{"id":"W2865216059","doi":"10.24963/ijcai.2018/787","title":"Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents (Extended Abstract)","year":2018,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Games","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Variety (cybernetics); Key (lock); Field (mathematics); Focus (optics); Data science; Human–computer interaction; Artificial intelligence; Management science; Computer security; Engineering","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":[],"consensus_categories":[],"category_scores_codex":[0.002288603,0.0001052029,0.0001034853,0.00002676604,0.0004889329,0.0006733649,0.0009594736,0.00003953359,0.0001788917],"category_scores_gemma":[0.0001987743,0.00007231638,0.00002700069,0.00007662437,0.0001090792,0.00063003,0.0005442108,0.0001041844,0.00007008998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004527638,"about_ca_system_score_gemma":0.00004087128,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004877123,"about_ca_topic_score_gemma":0.000006606703,"domain_scores_codex":[0.9986737,0.0001075068,0.0002845784,0.0003877966,0.0003028298,0.0002435791],"domain_scores_gemma":[0.9992421,0.0001231047,0.0001700675,0.0003292992,0.00008733677,0.00004814295],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001002926,0.00002180178,0.0007164352,0.00001972737,0.00001489644,3.29469e-7,0.002386968,0.001276291,0.003783128,0.0200927,0.0002611949,0.9714165],"study_design_scores_gemma":[0.000183499,0.0001772014,0.007806679,0.00006946284,0.000009028451,0.000004547244,0.0001405126,0.9006414,0.01847531,0.03074978,0.0415607,0.00018185],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05652042,0.00006514404,0.8546042,0.005489369,0.0001560578,0.06346489,0.000001544884,0.0001640929,0.01953424],"genre_scores_gemma":[0.875204,0.00001028792,0.1023365,0.0005386227,0.0006498464,0.01880462,0.00000391622,0.00002124891,0.002430947],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9712346,"threshold_uncertainty_score":0.6493273,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.157403217872697,"score_gpt":0.4028998136404876,"score_spread":0.2454965957677906,"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."}}