{"id":"W2953243993","doi":"10.48550/arxiv.1906.08226","title":"Unsupervised State Representation Learning in Atari","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke; Université de Montréal","funders":"","keywords":"Representation (politics); Computer science; Benchmark (surveying); Feature learning; Artificial intelligence; Generative grammar; Variety (cybernetics); Generative model; Encoder; Code (set theory); Machine learning; State (computer science); Encoding (memory)","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"],"consensus_categories":[],"category_scores_codex":[0.0003655507,0.0002881083,0.0003407348,0.0004596479,0.00008993852,0.0002038295,0.00176071,0.0002197269,0.0000326909],"category_scores_gemma":[0.0000955173,0.0003668293,0.0001393995,0.0007944341,0.00005286479,0.0006247039,0.002423078,0.001197237,0.0003278588],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003306898,"about_ca_system_score_gemma":0.0002416259,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004382021,"about_ca_topic_score_gemma":0.00003301225,"domain_scores_codex":[0.9976953,0.0003000443,0.0002968857,0.00112831,0.000162985,0.0004165146],"domain_scores_gemma":[0.9979779,0.0001778272,0.0003265896,0.001296765,0.0001223591,0.00009856124],"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":[0.00001543129,0.00001767423,0.06098548,0.00005363647,0.00002990783,0.0001481538,0.0006015915,0.9312325,0.00001822337,0.006507329,0.00004845509,0.000341595],"study_design_scores_gemma":[0.0005895116,0.00004945182,0.0114899,0.0001021775,0.00001434684,0.000001274727,0.0000850027,0.984358,0.00005512653,0.002529733,0.0003527762,0.0003727458],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2053628,0.00001571682,0.7898845,0.00006444737,0.0004888374,0.0003239986,0.000001109651,0.0002467693,0.003611836],"genre_scores_gemma":[0.9878013,0.0001726059,0.001850078,0.00005757562,0.00002445973,7.650584e-7,0.00003998687,0.00002176356,0.01003144],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7880344,"threshold_uncertainty_score":0.9998783,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0804774768729731,"score_gpt":0.2042313606317191,"score_spread":0.123753883758746,"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."}}