{"id":"W3127605640","doi":"10.1007/s00521-021-06375-y","title":"Improving reinforcement learning with human assistance: an argument for human subject studies with HIPPO Gym","year":2021,"lang":"en","type":"preprint","venue":"Neural Computing and Applications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure; University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; University of Alberta; Alberta Machine Intelligence Institute; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Amazon Web Services; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Computer science; Artificial intelligence; Human–computer interaction; Subject (documents); Parsing; World Wide Web","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":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.0004650394,0.0004746795,0.0005274272,0.0001346215,0.001895075,0.001011796,0.0007969283,0.0001187334,0.000001452468],"category_scores_gemma":[0.00002336085,0.0004046974,0.00008224559,0.000264222,0.0001535498,0.0002255339,0.001158517,0.0008609045,5.855324e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001632875,"about_ca_system_score_gemma":0.0001195298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006051272,"about_ca_topic_score_gemma":0.0000385089,"domain_scores_codex":[0.9971564,0.0001061487,0.0005530705,0.001217974,0.0004534983,0.0005129295],"domain_scores_gemma":[0.9974411,0.000189474,0.0007597557,0.001009352,0.0004526635,0.0001476905],"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":[0.00000719402,0.00003628207,0.001176854,0.0004660084,0.000131216,0.000003961053,0.001262728,0.9811525,0.0006907405,0.006926401,0.00001214976,0.00813399],"study_design_scores_gemma":[0.0009230411,0.001202197,0.00139241,0.0005301041,0.0001548038,0.00002448457,0.001277621,0.9921394,0.0006385593,0.0002804533,0.0005450806,0.0008918316],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1289441,0.00022714,0.8687284,0.000153243,0.00007308349,0.001266791,0.000001028857,0.0003997401,0.0002065259],"genre_scores_gemma":[0.9418553,0.00001674993,0.05675163,0.0001142876,0.0002265503,0.0004608188,0.0001598193,0.00004496305,0.0003698906],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8129112,"threshold_uncertainty_score":0.9998405,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04079369190192602,"score_gpt":0.3265417845465773,"score_spread":0.2857480926446513,"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."}}