{"id":"W3110181803","doi":"10.1609/aaai.v35i8.16820","title":"Decentralized Multi-Agent Linear Bandits with Safety Constraints","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"National Science Foundation","keywords":"Regret; Computer science; Mathematical optimization; Network topology; Upper and lower bounds; Bipartite graph; Linear programming; Graph; Telecommunications network; Mathematics; Theoretical computer science; Computer network; Machine learning","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001673607,0.0003089238,0.0005287943,0.0001876798,0.0003664207,0.0003509172,0.001861379,0.0001244315,0.00231163],"category_scores_gemma":[0.007015855,0.0001873495,0.0001884749,0.00175999,0.001313218,0.0003890403,0.0004482476,0.0005330491,0.0004723447],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000958323,"about_ca_system_score_gemma":0.0004848277,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001401201,"about_ca_topic_score_gemma":0.00005515578,"domain_scores_codex":[0.9948213,0.00008329978,0.00110297,0.0009181564,0.002443022,0.0006312521],"domain_scores_gemma":[0.9938836,0.0006833237,0.0005697558,0.0005372053,0.004070977,0.0002551886],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0024337,0.001747181,0.0051853,0.00008791428,0.0002033945,0.00006496065,0.004172777,0.003589701,0.10736,0.2528467,0.001055226,0.6212531],"study_design_scores_gemma":[0.0002780351,0.0002945799,0.001287637,0.0003134021,0.00002569158,0.000046888,0.004934984,0.04741533,0.8890674,0.05389563,0.002007994,0.0004324111],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3714171,0.0002921396,0.5810615,0.01734047,0.002229447,0.003067033,0.0002940631,0.0002795687,0.02401863],"genre_scores_gemma":[0.9874324,0.0001317736,0.01015152,0.0001726106,0.00007073767,0.00002210875,0.000002109139,0.00002347546,0.001993274],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7817074,"threshold_uncertainty_score":0.9986004,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2668979505252815,"score_gpt":0.4337449198338272,"score_spread":0.1668469693085457,"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."}}