文件名称:Message-Passing-Algorithms
- 所属分类:
- 图形图像处理(光照,映射..)
- 资源属性:
- [MacOS] [PDF]
- 上传时间:
- 2012-11-26
- 文件大小:
- 245kb
- 下载次数:
- 0次
- 提 供 者:
- fanha******
- 相关连接:
- 无
- 下载说明:
- 别用迅雷下载,失败请重下,重下不扣分!
介绍说明--下载内容均来自于网络,请自行研究使用
一本将基于AP传播算法的半监督聚类的算方法书.对于聚类研究的很有帮助-The Uncapacitated Facility Location Prob-
lem (UFLP) is one of the most widely stud-
ied discrete location problems, whose appli-
cations arise in a variety of settings. We
tackle the UFLP using probabilistic infer-
ence in a graphical model- an approach that
has received little attention in the past. We
show that the fi xed points of max-product
linear programming (MPLP), a convexifi ed
version of the max-product algorithm, can
be used to construct a solution with a 3-
approximation guarantee for metric UFLP
instances. In addition, we characterize some
scenarios under which the MPLP solution is
guaranteed to be globally optimal. We eval-
uate the performance of both max-sum and
MPLP empirically on metric and non-metric
problems, demonstrating the advantages of
the 3-approximation construction and algo-
rithm applicability to non-metric instances.
lem (UFLP) is one of the most widely stud-
ied discrete location problems, whose appli-
cations arise in a variety of settings. We
tackle the UFLP using probabilistic infer-
ence in a graphical model- an approach that
has received little attention in the past. We
show that the fi xed points of max-product
linear programming (MPLP), a convexifi ed
version of the max-product algorithm, can
be used to construct a solution with a 3-
approximation guarantee for metric UFLP
instances. In addition, we characterize some
scenarios under which the MPLP solution is
guaranteed to be globally optimal. We eval-
uate the performance of both max-sum and
MPLP empirically on metric and non-metric
problems, demonstrating the advantages of
the 3-approximation construction and algo-
rithm applicability to non-metric instances.
(系统自动生成,下载前可以参看下载内容)
下载文件列表
Message Passing Algorithms..pdf