{"id":"W4307490062","doi":"10.1613/jair.1.13854","title":"Low-Rank Representation of Reinforcement Learning Policies","year":2022,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Canadian Institute for Advanced Research; McGill University","funders":"Canadian Institute for Advanced Research","keywords":"Reproducing kernel Hilbert space; Reinforcement learning; Representation (politics); Computer science; Embedding; Rank (graph theory); Stability (learning theory); Kernel (algebra); Convergence (economics); Space (punctuation); Hilbert space; Hilbert curve; Mathematical optimization; Artificial intelligence; Theoretical computer science; Machine learning; Mathematics; Algorithm; Discrete mathematics; Economics; Pure mathematics","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":[0.00537001,0.0001074579,0.0002770267,0.001012713,0.000568799,0.0001770534,0.001710919,0.00003864954,0.0002914983],"category_scores_gemma":[0.001442927,0.0001038176,0.000169239,0.002077997,0.0001936349,0.0005507454,0.0009252095,0.001320764,0.00002958772],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002678323,"about_ca_system_score_gemma":0.0003894429,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001030655,"about_ca_topic_score_gemma":0.000001907063,"domain_scores_codex":[0.9945222,0.0008052091,0.001207599,0.0002023563,0.00278616,0.0004764741],"domain_scores_gemma":[0.9967926,0.000771342,0.0007210649,0.0004157782,0.001172203,0.0001270731],"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":[0.0000753425,0.00006900338,0.0002356925,0.0000195496,0.00003071508,0.00002714229,0.003841818,0.9211692,0.01193041,0.0353643,0.0002415951,0.02699524],"study_design_scores_gemma":[0.00007576452,0.002514604,0.0001204099,0.00006638232,0.000007987888,0.0000732461,0.008277825,0.7822275,0.1972213,0.007309578,0.001944637,0.0001607353],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04167698,0.00007808381,0.9548143,0.001123031,0.0004668419,0.0001879355,2.610448e-7,0.00001875455,0.001633806],"genre_scores_gemma":[0.995194,0.0001309306,0.003857214,0.00002969318,0.0001453157,0.000007332543,0.000001077391,0.0000109489,0.0006234684],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.953517,"threshold_uncertainty_score":0.5738138,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1582119241585682,"score_gpt":0.4211595401058433,"score_spread":0.2629476159472751,"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."}}