{"id":"W2724228633","doi":"10.1145/3071178.3071303","title":"Multi-task learning in Atari video games with emergent tangled program graphs","year":2017,"lang":"en","type":"article","venue":"Proceedings of the Genetic and Evolutionary Computation Conference","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Observability; Task (project management); Video game; Artificial intelligence; Variety (cybernetics); State (computer science); Matching (statistics); Human–computer interaction; Machine learning; Multimedia; Programming language","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":[],"consensus_categories":[],"category_scores_codex":[0.0001702776,0.0001484724,0.0001639612,0.0000952959,0.0004408393,0.0002351451,0.0008205245,0.00004830882,0.000003073558],"category_scores_gemma":[0.0001208143,0.0001136503,0.00003550985,0.000179695,0.0002545675,0.0004232615,0.0004553265,0.000187934,0.000002148038],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000279621,"about_ca_system_score_gemma":0.00009073815,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005210789,"about_ca_topic_score_gemma":0.000006573462,"domain_scores_codex":[0.9988236,0.00001651692,0.0002798279,0.0003349307,0.0003241078,0.0002209996],"domain_scores_gemma":[0.9989178,0.00003009475,0.0004547024,0.0001684134,0.0003712839,0.00005769295],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003391296,0.0001154596,0.6290751,0.0001344063,0.00004392481,0.000001544954,0.001839463,0.3250223,0.001127561,0.007633473,0.0001709395,0.03480201],"study_design_scores_gemma":[0.0003295328,0.0001254966,0.4577007,0.00008622028,0.000006566867,0.000008311287,0.0000689423,0.5403116,0.00004892419,0.001117837,0.0001028693,0.00009300018],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5626723,0.0001705625,0.4345628,0.001060645,0.000157999,0.0006107312,6.313194e-7,0.0001081186,0.000656299],"genre_scores_gemma":[0.8442615,0.0000604423,0.1554326,0.000017801,0.00001014573,0.00002979217,0.000001083106,0.000006284657,0.000180367],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2815892,"threshold_uncertainty_score":0.4634525,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02192359355186942,"score_gpt":0.2563764809461951,"score_spread":0.2344528873943257,"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."}}