1. NCCU Academic Hub
  2. 學術產出
  3. College of Informatics
  4. Periodical Articles
  5. Item

學術產出-Periodical Articles

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

題名 On state-space abstraction for anytime evaluation of Bayesian networks
作者 Liu, Chao-lin;Wellman, Michael P.
劉昭麟
貢獻者 資科系
日期 1996
上傳時間 17-Jun-2015 15:07:54 (UTC+8)
摘要 Despite the increasing popularity of Bayesian networks for representing and reasoning about uncertain situations, the complexity of inference in this formalism remains a significant concern. A viable approach to relieving the problem is trading off accuracy for computational efficiency. To this end, varying the granularity of state space of state variables appears to be a feasible strategy for controlling the evaluation process. We consider an anytime procedure for approximate evaluation of Bayesian networks based on this idea. On application to some simple networks, the procedure exhibits a smooth improvement in approximation quality as computation time increases. With the aim of developing principled control techniques, we also conduct a theoretical analysis of the quality of approximation. Our main result demonstrates that the error induced by state-space abstraction deceases with the distance from the abstracted nodes, where "distance" is defined in terms of d-separation. While the empirical results suggest that incremental state-space abstraction offers a viable performance profile, the theoretical results represent a starting point for the deliberation scheduling of our anytime approximation method.
關聯 Intelligence/sigart Bulletin - SIGART , vol. 7, no. 2, pp. 50-57
資料類型 article
DOI http://dx.doi.org/10.1145/242587.242601
dc.contributor 資科系
dc.creator (作者) Liu, Chao-lin;Wellman, Michael P.
dc.creator (作者) 劉昭麟zh_TW
dc.date (日期) 1996
dc.date.accessioned 17-Jun-2015 15:07:54 (UTC+8)-
dc.date.available 17-Jun-2015 15:07:54 (UTC+8)-
dc.date.issued (上傳時間) 17-Jun-2015 15:07:54 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/75885-
dc.description.abstract (摘要) Despite the increasing popularity of Bayesian networks for representing and reasoning about uncertain situations, the complexity of inference in this formalism remains a significant concern. A viable approach to relieving the problem is trading off accuracy for computational efficiency. To this end, varying the granularity of state space of state variables appears to be a feasible strategy for controlling the evaluation process. We consider an anytime procedure for approximate evaluation of Bayesian networks based on this idea. On application to some simple networks, the procedure exhibits a smooth improvement in approximation quality as computation time increases. With the aim of developing principled control techniques, we also conduct a theoretical analysis of the quality of approximation. Our main result demonstrates that the error induced by state-space abstraction deceases with the distance from the abstracted nodes, where "distance" is defined in terms of d-separation. While the empirical results suggest that incremental state-space abstraction offers a viable performance profile, the theoretical results represent a starting point for the deliberation scheduling of our anytime approximation method.
dc.format.extent 838111 bytes-
dc.format.mimetype application/pdf-
dc.relation (關聯) Intelligence/sigart Bulletin - SIGART , vol. 7, no. 2, pp. 50-57
dc.title (題名) On state-space abstraction for anytime evaluation of Bayesian networks
dc.type (資料類型) articleen
dc.identifier.doi (DOI) 10.1145/242587.242601
dc.doi.uri (DOI) http://dx.doi.org/10.1145/242587.242601