{"id":"W3164770271","doi":"10.1007/s00521-021-06104-5","title":"Lucid dreaming for experience replay: refreshing past states with the current policy","year":2021,"lang":"en","type":"article","venue":"Neural Computing and Applications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"Army Research Office; Defense Advanced Research Projects Agency; Office of Naval Research; Future of Life Institute; Alberta Machine Intelligence Institute; Natural Sciences and Engineering Research Council of Canada; Lockheed Martin; Robert Bosch (Australia) Pty; Canadian Institute for Advanced Research; National Science Foundation","keywords":"Computer science; State (computer science); Reinforcement learning; Work (physics); Dream; Artificial intelligence; Programming language; Psychology","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":[0.0001286021,0.0001032719,0.0000912933,0.00003267394,0.0007677443,0.000369718,0.0003869843,0.00001448378,2.690077e-7],"category_scores_gemma":[0.00003515917,0.00007380029,0.00002601285,0.0003957428,0.00006834753,0.0001273974,0.0002546616,0.0001679704,0.000001360228],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002040451,"about_ca_system_score_gemma":0.00005155434,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001164481,"about_ca_topic_score_gemma":0.000001741106,"domain_scores_codex":[0.9990786,0.0000336669,0.0001591238,0.0003463282,0.0001476126,0.0002346912],"domain_scores_gemma":[0.998926,0.0003352689,0.0001145958,0.0004505797,0.0001171444,0.00005638801],"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":[0.000004018198,0.00003101252,0.001395577,0.00005825404,0.00001522835,0.000001964862,0.00485848,0.4313225,0.0005592902,0.06989637,0.000709064,0.4911483],"study_design_scores_gemma":[0.0001673586,0.00004190386,0.001081146,0.00003470333,0.00000615434,0.00003213072,0.0003145959,0.934244,0.0005855259,0.0003778464,0.06296715,0.0001474318],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02403066,0.0001717366,0.9707156,0.004501353,0.00004676335,0.0002474944,0.000001279037,0.0001228607,0.0001622513],"genre_scores_gemma":[0.9753448,0.0000198347,0.02357493,0.0005722154,0.0002714881,0.00008271238,0.00001146054,0.000009512303,0.0001130705],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9513141,"threshold_uncertainty_score":0.5904946,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01889025861605372,"score_gpt":0.3080537549947229,"score_spread":0.2891634963786692,"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."}}