.png){kind=icon}
返回
Approximate inference in Bayesian networks: Parameterized complexity results
.png){kind=icon}
OA
AI
DOI：10.1016/j.ijar.2017.10.029.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Computing posterior and marginal probabilities constitutes the backbone of almost all inferences in Bayesian networks. These computations are known to be intractable in general, both to compute exactly and to approximate (e.g., by sampling algorithms). While it is well known under what constraints exact computation can be rendered tractable (viz., bounding tree-width of the moralized network and bounding the cardinality of the variables) it is less known under what constraints approximate Bayesian inference can be tractable. Here, we extend the existing formal framework of fixed-error randomized tractability (a randomized analogue of fixed-parameter tractability), and use it to address this problem, both by re-interpreting known results from the literature and by providing some additional new results, including results on fixed parameter tractable de-randomization of approximate inference. (C) 2017 Elsevier Inc. All rights reserved.
Keyword:
Approximate inference
Sampling
Bayesian networks
Parameterized complexity
Stochastic algorithms
De-randomization
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3
论文数:
3.0K
被引数:
5.1K
机构
引用论文
.png){kind=icon}
A note on variability of water extract viscosity of
rye grain from north-east regions of Poland关于波兰东北部地区黑麦谷物水提取液粘度变异性的注释
OAAI
OAAI分享收藏
查看更多内容