{"id":"W1904406446","doi":"","title":"Using predictive representations to improve generalization in reinforcement learning","year":2005,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Generalization; Reinforcement learning; Computer science; Representation (politics); Artificial intelligence; Task (project management); State (computer science); Machine learning; Grid; Reinforcement; Algorithm; Mathematics; Psychology; Social psychology; Engineering","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":[],"consensus_categories":[],"category_scores_codex":[0.000199068,0.0001062941,0.00009891468,0.0002514132,0.0001153684,0.000139075,0.0003435168,0.00004089949,0.00004722322],"category_scores_gemma":[0.000133338,0.0001098743,0.00002338161,0.0005991648,0.00001154711,0.0007816578,0.0002706987,0.0001350812,0.00006570664],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002444497,"about_ca_system_score_gemma":0.00005630468,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001288032,"about_ca_topic_score_gemma":0.00001775189,"domain_scores_codex":[0.9987137,0.00005801034,0.000335251,0.0003161243,0.000304632,0.0002722787],"domain_scores_gemma":[0.9993632,0.00003814775,0.00009188029,0.0003345603,0.00008932288,0.00008282709],"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":[0.000002707687,0.000007419376,0.00166425,0.000002302456,0.000005347736,9.976446e-7,0.001145942,0.9805189,0.003754505,0.01121536,0.0001003163,0.001581974],"study_design_scores_gemma":[0.0002371877,0.000100149,0.001082736,0.00001257112,0.000002692937,0.000001611131,0.00006939652,0.9936538,0.003253779,0.00003560127,0.001423946,0.0001264759],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005559849,0.000004877233,0.9814273,0.0005219124,0.0001456693,0.0003224565,7.701384e-8,0.0001518201,0.01186601],"genre_scores_gemma":[0.7642272,0.000004438094,0.2308096,0.0004773896,0.00008318221,0.00001996685,0.000004481071,0.000008932482,0.004364863],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7586673,"threshold_uncertainty_score":0.4480545,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02994124902249506,"score_gpt":0.3081307393099424,"score_spread":0.2781894902874474,"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."}}