{"id":"W1799762961","doi":"10.1613/jair.904","title":"Accelerating Reinforcement Learning by Composing Solutions of Automatically Identified Subtasks","year":2002,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":54,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada; Government of Ontario","keywords":"Reinforcement learning; Exploit; Partition (number theory); Function (biology); Base (topology); State space; Reinforcement; Function approximation","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":[0.005513628,0.0001582135,0.0003574942,0.0006347562,0.0006490577,0.0006008826,0.001532451,0.0001035378,0.0002863946],"category_scores_gemma":[0.00198227,0.0001477241,0.0001695905,0.001206118,0.0003029009,0.0009707657,0.0004861951,0.001340429,0.0001176081],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002024191,"about_ca_system_score_gemma":0.0001393467,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003391218,"about_ca_topic_score_gemma":0.000001860025,"domain_scores_codex":[0.9945855,0.0005479954,0.001741448,0.000238769,0.002137171,0.0007491728],"domain_scores_gemma":[0.9959714,0.001067399,0.000760043,0.0004041972,0.001574001,0.0002229691],"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":[0.00001844722,0.0001522497,0.00008921423,0.00005481222,0.00006559226,0.00002636038,0.002875281,0.840547,0.07357518,0.0358612,0.001861739,0.04487291],"study_design_scores_gemma":[0.00004797426,0.0006240203,0.00001583854,0.0001882794,0.000008033384,0.00003872816,0.0006874988,0.9460827,0.05035247,0.001466805,0.0003590491,0.0001286611],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01075093,0.000221886,0.9853038,0.001087945,0.0003047289,0.0001803519,2.653346e-7,0.00003623838,0.002113923],"genre_scores_gemma":[0.9750773,0.0001424225,0.02402497,0.00002192062,0.0001155593,0.00000276163,9.130298e-7,0.00001436146,0.0005998358],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9643263,"threshold_uncertainty_score":0.6024014,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2748869913500262,"score_gpt":0.396747064427609,"score_spread":0.1218600730775828,"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."}}