{"id":"W4393277074","doi":"10.54254/2755-2721/54/20241586","title":"Assessing the robustness of Multi-Armed Bandit algorithms against biased initialization","year":2024,"lang":"en","type":"article","venue":"Applied and Computational Engineering","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Regret; Initialization; Robustness (evolution); Adaptability; Computer science; Recommender system; Thompson sampling; Greedy algorithm; Upper and lower bounds; Commit; Machine learning; Algorithm; Mathematical optimization; Artificial intelligence; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008144965,0.0001143358,0.0001588503,0.0002660641,0.0001346779,0.0004249678,0.0002114103,0.00004382944,0.00001825421],"category_scores_gemma":[0.0002524476,0.00007858108,0.00004033342,0.0008465434,0.00008357991,0.0002950849,0.0000802638,0.0001552462,0.000007810107],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002649854,"about_ca_system_score_gemma":0.00006199832,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001674283,"about_ca_topic_score_gemma":5.873196e-7,"domain_scores_codex":[0.9983503,0.00002890281,0.000364312,0.000285249,0.0008091539,0.0001620799],"domain_scores_gemma":[0.9976436,0.001966796,0.00005536381,0.0001231298,0.0001574132,0.00005368242],"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":[0.000001770461,0.00001007879,0.000007576039,0.00001918419,0.00001529921,0.000003645149,0.0001467869,0.9321075,0.0003492101,0.001731449,0.00002617342,0.06558134],"study_design_scores_gemma":[0.0002237539,0.000006012438,0.002071497,0.00003511054,0.000004712531,0.000005795252,0.0002456867,0.9949982,0.0002772828,0.00146373,0.000574765,0.00009344728],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02150326,0.0003160054,0.9774683,0.0001049424,0.0002659006,0.0001544523,0.00001053314,0.00006848317,0.0001081642],"genre_scores_gemma":[0.9756388,0.00001087527,0.02408961,0.00001930051,0.0001505903,0.00002338973,0.00002845306,0.0000170294,0.00002196774],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9541355,"threshold_uncertainty_score":0.4097975,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.101806615478968,"score_gpt":0.405559755997657,"score_spread":0.303753140518689,"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."}}