{"id":"W2594903727","doi":"10.48550/arxiv.1702.08360","title":"Neural Map: Structured Memory for Deep Reinforcement Learning","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":102,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Computer science; Observability; Artificial intelligence; Set (abstract data type); Artificial neural network; Simple (philosophy); Architecture; Memory map; Deep learning; Shared memory; Parallel computing; Programming language","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":[0.0004211208,0.0006833353,0.0005549386,0.0002509948,0.0002019836,0.0006200006,0.001567157,0.0008497731,0.00432108],"category_scores_gemma":[0.002068703,0.0003187584,0.0003738887,0.0002399089,0.0005361558,0.001038834,0.001092896,0.00141401,0.000771816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005590847,"about_ca_system_score_gemma":0.0006332793,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002570417,"about_ca_topic_score_gemma":0.003272894,"domain_scores_codex":[0.9998507,0.00003380147,0.0000081632,0.00004336777,0.00004166451,0.00002218956],"domain_scores_gemma":[0.9996301,0.0001652534,0.0000378593,0.00006660152,0.00006757618,0.00003257341],"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":[0.0002085472,0.0001057428,0.0008845913,0.0001495331,0.00007889429,0.0001454301,0.00007615759,0.7823646,0.006777955,0.02809767,0.005828615,0.1752823],"study_design_scores_gemma":[0.000008890415,0.0000265073,0.0000491022,0.000004349004,0.000003844474,0.00001056057,0.000003590888,0.9876961,0.001263774,0.01024227,0.0006877733,0.000003349943],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01888483,0.000324874,0.9746667,0.0002886176,0.00009025752,0.00004001774,0.0001967086,0.002647771,0.002860288],"genre_scores_gemma":[0.7949371,0.0002878614,0.1985422,0.0002054427,0.00004962287,0.0002409068,0.0003649845,0.0001750644,0.005196866],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00432108,"threshold_uncertainty_score":0.01445544,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0666945800872272,"score_gpt":0.2102999486985228,"score_spread":0.1436053686112956,"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."}}