{"id":"W2963097726","doi":"10.1609/aaai.v33i01.33014384","title":"The Utility of Sparse Representations for Control in Reinforcement Learning","year":2019,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Sparse approximation; Computer science; Neural coding; Bootstrapping (finance); Artificial intelligence; Representation (politics); Machine learning; Artificial neural network; Locality; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002574991,0.0007978194,0.0009166682,0.0004942686,0.0004284583,0.00119405,0.001078513,0.001298827,0.002172578],"category_scores_gemma":[0.01558663,0.0004123802,0.0006348696,0.0004909488,0.00255423,0.00278794,0.00149394,0.002747966,0.0002706471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009452344,"about_ca_system_score_gemma":0.0008674151,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00178732,"about_ca_topic_score_gemma":0.001221239,"domain_scores_codex":[0.9989793,0.000507112,0.00004605172,0.0001651807,0.0002121189,0.00009024705],"domain_scores_gemma":[0.9930421,0.005407297,0.0004770858,0.0005513339,0.0003429693,0.0001792103],"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":[0.0001073266,0.00008588642,0.0007096431,0.0001271371,0.0000389486,0.00008694277,0.0001542058,0.6488802,0.002596839,0.3052908,0.000901512,0.04102064],"study_design_scores_gemma":[0.00001417701,0.00006371627,0.00005358831,0.00001118798,0.00000550539,0.00001311834,0.000007717714,0.910542,0.0004954935,0.08841015,0.0003758443,0.000007458019],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01388713,0.0002069428,0.9828636,0.0005426632,0.00003160598,0.00003503421,0.00002903536,0.00009669245,0.002307247],"genre_scores_gemma":[0.8551536,0.0004919745,0.1411837,0.0003840328,0.0001289182,0.0001940583,0.00006512964,0.0000676764,0.002330898],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002574991,"threshold_uncertainty_score":0.01361805,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06591031971543576,"score_gpt":0.3089931647799222,"score_spread":0.2430828450644865,"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."}}