| dc.contributor | 統計系 | en_US |
| dc.creator (作者) | 洪英超 | zh_TW |
| dc.date (日期) | 2012.01 | en_US |
| dc.date.accessioned | 11-Nov-2013 17:42:24 (UTC+8) | - |
| dc.date.available | 11-Nov-2013 17:42:24 (UTC+8) | - |
| dc.date.issued (上傳時間) | 11-Nov-2013 17:42:24 (UTC+8) | - |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/61596 | - |
| dc.description.abstract (摘要) | We consider the bandit problem with an infinite number of Bernoulli arms, of which the unknown parameters are assumed to be i.i.d. random variables with a common distribution F. Our goal is to construct optimal strategies of choosing “arms” so that the expected long-run failure rate is minimized. We first review a class of strategies and establish their asymptotic properties when F is known. Based on the results, we propose a new strategy and prove that it is asymptotically optimal when F is unknown. Finally, we show that the proposed strategy performs well for a number of simulation scenarios. | en_US |
| dc.format.extent | 197602 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.language.iso | en_US | - |
| dc.relation (關聯) | Journal of Statistical Planning and Inference, 142, 86-94 | en_US |
| dc.subject (關鍵詞) | Bandit problem;Bernoulli arms;Bayesian strategy;Prior distribution | en_US |
| dc.title (題名) | Optimal Bayesian Strategies for the Infinite-armed Bernoulli Bandit | en_US |
| dc.type (資料類型) | article | en |
| dc.identifier.doi (DOI) | 10.1016/j.jspi.2011.06.026 | en_US |
| dc.doi.uri (DOI) | http://dx.doi.org/http://dx.doi.org/10.1016/j.jspi.2011.06.026 | en_US |
