{"id":"W4405254983","doi":"10.48550/arxiv.2412.07165","title":"A Method for Evaluating Hyperparameter Sensitivity in Reinforcement Learning","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Evolutionary Algorithms and Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Alberta Machine Intelligence Institute; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Hyperparameter; Sensitivity (control systems); Reinforcement learning; Reinforcement; Machine learning; Artificial intelligence; Psychology; Computer science; Social psychology; Engineering","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001146391,0.0002034459,0.0002368798,0.0002427399,0.0001468424,0.00009800041,0.0004566425,0.0001599382,0.000005866108],"category_scores_gemma":[0.00005231089,0.0002386425,0.0001560931,0.000519465,0.00002979109,0.000145191,0.001784837,0.000659482,0.00004054893],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002795612,"about_ca_system_score_gemma":0.0002091939,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002019332,"about_ca_topic_score_gemma":0.00003647507,"domain_scores_codex":[0.9982936,0.0001730941,0.0002024529,0.0009650575,0.00008480725,0.0002809757],"domain_scores_gemma":[0.9987566,0.0003790667,0.00013205,0.0005367608,0.000124544,0.00007097481],"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":[0.000005943477,0.00002487726,0.00009553204,0.00006831711,0.00002531495,0.00003219687,0.0001682354,0.836042,0.0001317092,0.1598217,0.0000382506,0.003545829],"study_design_scores_gemma":[0.0002032802,0.00005280784,0.0002077985,0.00007346685,0.00003837905,0.000004797168,0.00004073999,0.8873249,0.00005825658,0.1114395,0.0003269156,0.000229194],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04695495,0.00004490843,0.9510199,0.0002597783,0.0001756649,0.0005404193,0.000004988212,0.0001665664,0.0008327884],"genre_scores_gemma":[0.8049072,0.00001373902,0.1933194,0.00004859841,0.00004572531,0.00001353963,0.00001600805,0.00001168942,0.001624129],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7579523,"threshold_uncertainty_score":0.9731559,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1114899556047503,"score_gpt":0.2707874725826567,"score_spread":0.1592975169779064,"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."}}