{"id":"W2794940174","doi":"10.1609/aimag.v39i1.2780","title":"Constructing Temporal Abstractions Autonomously in Reinforcement Learning","year":2018,"lang":"en","type":"article","venue":"AI Magazine","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Computer science; Abstraction; Artificial intelligence; Thread (computing); Architecture; 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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0003928268,0.0001580497,0.0001688405,0.000223987,0.0001878177,0.0002338862,0.0004886997,0.00006733733,0.0002472234],"category_scores_gemma":[0.0001252402,0.0001670216,0.00003769017,0.0004150989,0.000120899,0.0007095125,0.0002655679,0.0004472286,0.001426565],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001728766,"about_ca_system_score_gemma":0.0001139843,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007467556,"about_ca_topic_score_gemma":0.00005668319,"domain_scores_codex":[0.998492,0.00004707943,0.0004477561,0.0003272219,0.0002700034,0.0004159285],"domain_scores_gemma":[0.9991171,0.00007608535,0.0001921247,0.000411153,0.0001206033,0.00008294012],"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":[0.00001161914,0.00003613886,0.05936127,0.00002546833,0.00003394283,0.00005680373,0.001222014,0.829207,0.001060829,0.08367407,0.002598504,0.02271236],"study_design_scores_gemma":[0.000554037,0.0003467412,0.005861467,0.00004276499,0.000004785039,0.00005264407,0.000098647,0.9375591,0.0005179968,0.0002533033,0.05442018,0.0002882986],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007318405,0.000007668798,0.9083323,0.001369257,0.0005894771,0.0001607008,1.391278e-7,0.0003065499,0.08191546],"genre_scores_gemma":[0.962129,0.000002748995,0.03106007,0.000431238,0.0001531367,0.000008865896,0.000008388457,0.00001315925,0.00619337],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9548106,"threshold_uncertainty_score":0.999351,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01520819445972775,"score_gpt":0.2693107594008761,"score_spread":0.2541025649411484,"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."}}