{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001145652,0.0009350267,0.0007912785,0.0004615308,0.0003441231,0.0005655198,0.001543995,0.00071445,0.003616001],"category_scores_gemma":[0.004633822,0.0004698567,0.0004359708,0.0002979573,0.0006059353,0.001242458,0.001217726,0.001171402,0.001070484],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004460455,"about_ca_system_score_gemma":0.0006759282,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001485715,"about_ca_topic_score_gemma":0.001297837,"domain_scores_codex":[0.9995541,0.0001074651,0.00003204748,0.0001264772,0.0001287145,0.00005118828],"domain_scores_gemma":[0.9984424,0.0007934638,0.0001889012,0.0002841071,0.0001915959,0.00009956884],"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":[0.0005961486,0.000903324,0.00233347,0.0001933537,0.0001075187,0.0001883825,0.0003313167,0.3132252,0.06351679,0.006676112,0.001691532,0.6102368],"study_design_scores_gemma":[0.00009314164,0.0002727565,0.0004715872,0.00001451725,0.00003611137,0.00009134388,0.00002689647,0.9674308,0.02352276,0.005551459,0.0024683,0.00002031397],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09591879,0.00008921867,0.8962111,0.0000960365,0.00004728784,0.0002545096,0.00002540708,0.004415188,0.002942406],"genre_scores_gemma":[0.5686299,0.00009953748,0.4271002,0.00008123036,0.00002974094,0.0003129284,0.0001161823,0.0002664943,0.003363631],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003616001,"threshold_uncertainty_score":0.01209676,"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."}}