{"id":"W7071380059","doi":"","title":"Risk-directed exploration in reinforcement learning","year":2005,"lang":"en","type":"dissertation","venue":"eScholarship@McGill (McGill)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Reinforcement learning; Entropy (arrow of time); Measure (data warehouse); Action (physics); Reinforcement; Action selection; Class (philosophy)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001773551,0.001002509,0.0008700825,0.001217411,0.001231529,0.0004143504,0.002013813,0.0009229518,0.0001785733],"category_scores_gemma":[0.001841502,0.001161986,0.0002809724,0.001691458,0.00003715168,0.003569243,0.0003918744,0.003777923,0.0009379215],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001584541,"about_ca_system_score_gemma":0.0001248961,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003024747,"about_ca_topic_score_gemma":0.001080177,"domain_scores_codex":[0.9930285,0.0008335284,0.001741733,0.001573216,0.001656981,0.001166061],"domain_scores_gemma":[0.9957059,0.0003487638,0.001749108,0.001367893,0.0005055064,0.0003228196],"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":[0.00007935349,0.00008175991,0.00009975018,0.0001337552,0.000113061,0.00005276649,0.00009766249,0.8261395,0.002022094,0.04942349,0.000008722116,0.1217481],"study_design_scores_gemma":[0.00585928,0.001725478,0.006604472,0.002595559,0.0003821944,0.00004080585,0.001196282,0.7170789,0.04719104,0.01140202,0.198804,0.007120008],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.4276886,0.0005948108,0.005004959,0.0001114422,0.007656147,0.004507554,0.00006420284,0.006283541,0.5480887],"genre_scores_gemma":[0.9556458,0.001037904,0.01073479,0.0001570363,0.0001026316,0.0002747626,0.001164728,0.0002071419,0.0306752],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5279572,"threshold_uncertainty_score":0.99984,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01879923180380377,"score_gpt":0.2485763170795089,"score_spread":0.2297770852757051,"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."}}