{"id":"W2964278219","doi":"","title":"Thompson Sampling for Combinatorial Semi-Bandits.","year":2018,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Regret; Thompson sampling; Matroid; Oracle; Sampling (signal processing); Mathematics; Mathematical optimization; Independence (probability theory); Computer science; Combinatorics; Statistics","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":[0.004932829,0.001329104,0.001772542,0.001130018,0.000840617,0.001929459,0.001966929,0.001775364,0.005039861],"category_scores_gemma":[0.02216731,0.0007568907,0.001262768,0.001627303,0.002033267,0.002528863,0.001469856,0.002433606,0.0007680389],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00219685,"about_ca_system_score_gemma":0.001504463,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003290706,"about_ca_topic_score_gemma":0.003993835,"domain_scores_codex":[0.9973229,0.00171204,0.00009048361,0.0003019925,0.0003860582,0.0001865938],"domain_scores_gemma":[0.9865588,0.01099243,0.0008407296,0.0008237517,0.0004154668,0.0003687042],"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":[0.0002277232,0.000116461,0.001390554,0.0001910553,0.0001119666,0.0001210758,0.00008628379,0.6585526,0.0008958898,0.3043033,0.002969212,0.03103388],"study_design_scores_gemma":[0.0000189951,0.0000287394,0.00008692448,0.00001424134,0.000009335402,0.00001923344,0.000008611239,0.9270403,0.0001964904,0.07195295,0.0006186052,0.000005671583],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01752963,0.0007974888,0.9756141,0.0004623926,0.00009966984,0.0001342912,0.00009075631,0.0003003188,0.004971364],"genre_scores_gemma":[0.6492397,0.001086427,0.340095,0.0006281554,0.0002901192,0.00072301,0.0004263252,0.0002383651,0.007272953],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005039861,"threshold_uncertainty_score":0.02608764,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2722630354409833,"score_gpt":0.5115727209194741,"score_spread":0.2393096854784908,"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."}}