{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01861695,0.0009819345,0.001195144,0.001180398,0.001082284,0.002688934,0.001474091,0.001958963,0.0009896467],"category_scores_gemma":[0.1154298,0.0004394115,0.000650099,0.001109996,0.001626949,0.002415258,0.001738739,0.002074798,0.0003376411],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001497974,"about_ca_system_score_gemma":0.002100771,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007180152,"about_ca_topic_score_gemma":0.004381525,"domain_scores_codex":[0.9926341,0.004584461,0.0004125736,0.0008705436,0.001018929,0.0004792899],"domain_scores_gemma":[0.9033976,0.07846125,0.00665335,0.007181881,0.003279905,0.001026154],"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":[0.0005553152,0.0001579101,0.01691798,0.0001544315,0.0003250373,0.00009126173,0.0002256119,0.9425053,0.001280367,0.01245079,0.0007491506,0.02458693],"study_design_scores_gemma":[0.00005037904,0.0004448721,0.00346997,0.00006782164,0.00006293137,0.00007214634,0.0001875103,0.9828471,0.001661514,0.01053652,0.0005647863,0.00003448203],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7543615,0.001823122,0.233784,0.001548306,0.0001811504,0.0002831206,0.0003172499,0.0003958444,0.007305656],"genre_scores_gemma":[0.9743468,0.0003103731,0.02434392,0.0001587557,0.00003616298,0.0001129552,0.0001599454,0.000037188,0.0004939403],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01861695,"threshold_uncertainty_score":0.09845698,"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."}}