{"id":"W2904232271","doi":"10.1609/aaai.v33i01.33016112","title":"Leveraging Observations in Bandits: Between Risks and Benefits","year":2019,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Regret; Multi-armed bandit; Leverage (statistics); Computer science; Optimism; Context (archaeology); Thompson sampling; Order (exchange); Task (project management); Artificial intelligence; Machine learning; Dependency (UML); Economics; Psychology","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":[0.01239889,0.002015115,0.001993821,0.0007186384,0.0009622455,0.002383128,0.002050094,0.002793589,0.001443565],"category_scores_gemma":[0.07663623,0.001135382,0.0008565105,0.0005897645,0.004392011,0.005565762,0.004619733,0.005007812,0.0003485219],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001468588,"about_ca_system_score_gemma":0.001350405,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001736094,"about_ca_topic_score_gemma":0.00136341,"domain_scores_codex":[0.9942436,0.003374388,0.0002542742,0.0007499805,0.000978374,0.0003993816],"domain_scores_gemma":[0.9244561,0.06376859,0.00489589,0.005063485,0.001024729,0.0007912805],"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":[0.0005430088,0.0001447918,0.004004012,0.0001333124,0.0001238902,0.0001771138,0.0002506721,0.873076,0.001980441,0.08216585,0.000460153,0.03694078],"study_design_scores_gemma":[0.00004011356,0.0001512682,0.0005103769,0.00003799945,0.00002670692,0.00005509858,0.00002482188,0.937046,0.001061241,0.06074887,0.0002749446,0.00002261856],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1250243,0.001020988,0.8686066,0.001507365,0.00004574777,0.00008811346,0.0000503036,0.000505244,0.003151346],"genre_scores_gemma":[0.9427553,0.0003983952,0.05506595,0.0002371293,0.00006087906,0.0001344256,0.00003592786,0.00007910468,0.001232906],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01239889,"threshold_uncertainty_score":0.06557232,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.534976651882224,"score_gpt":0.4463792313613104,"score_spread":0.08859742052091357,"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."}}