{"id":"W3007722162","doi":"10.1109/tsp.2021.3089822","title":"Safe Linear Thompson Sampling With Side Information","year":2021,"lang":"en","type":"preprint","venue":"IEEE Transactions on Signal Processing","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Office of the President, University of California","keywords":"Frequentist inference; Regret; Thompson sampling; Set (abstract data type); Reliability (semiconductor); Linear programming; Mathematics; Computer science; Algorithm; Mathematical optimization; Sampling (signal processing); Combinatorics; Bayesian probability; Statistics; Bayesian inference","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":["metaepi_narrow","scholarly_communication","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001684147,0.000517416,0.0007027382,0.001083909,0.0008616734,0.002070806,0.0009979906,0.0004255416,0.0006525006],"category_scores_gemma":[0.0001403704,0.0004020096,0.0002633616,0.001592509,0.0002239795,0.002514508,0.00002709608,0.002437194,0.0002236003],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003084104,"about_ca_system_score_gemma":0.001436026,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004058522,"about_ca_topic_score_gemma":0.00007476027,"domain_scores_codex":[0.9931484,0.0002449434,0.001225366,0.00100141,0.003765423,0.0006144333],"domain_scores_gemma":[0.9952439,0.0009989151,0.0006585888,0.0008145927,0.002020287,0.0002636955],"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":[0.0001387385,0.00006062032,0.00000356161,0.00008122515,0.00002643734,0.00001253183,0.0007439232,0.5628438,0.0001622851,5.746999e-7,0.00000972403,0.4359166],"study_design_scores_gemma":[0.001585715,0.0004235718,0.0002027811,0.00251355,0.0001532621,0.0001467133,0.006140043,0.9164271,0.06202101,0.003205858,0.005483881,0.001696503],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005266611,0.0001854466,0.9919874,0.0004138884,0.0004458148,0.0005049622,0.00007977417,0.0001720861,0.0009440063],"genre_scores_gemma":[0.9170092,0.0000425475,0.08169092,0.0002018745,0.000215439,0.0001198266,0.00003606677,0.0000541179,0.0006299636],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9117426,"threshold_uncertainty_score":0.9998642,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1412922997182597,"score_gpt":0.4176311001368637,"score_spread":0.276338800418604,"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."}}