高斯信念传播的轨道 - 乘积表示与校正Orbit-Product Representation and Correction of Gaussian Belief Propagation |
|
课程网址: | http://videolectures.net/icml09_johnson_opr/ |
主讲教师: | Jason K. Johnson |
开课单位: | 麻省理工学院 |
开课时间: | 2009-08-26 |
课程语种: | 英语 |
中文简介: | 我们提出了一种新的高斯信念传播(GaBP)视图,它基于行列式的表示作为图形的乘积或位。我们证明了GaBP确定估计捕获了图的完全回溯轨道,并考虑如何纠正这个估计。我们证明了丢失轨道可以被分组为对应于无轨道轨道的等效类,并且可以从GaBP解决方案容易地确定每个等效类的贡献。此外,我们证明这个乘法校正因子可以解释为基于GaBP的边缘权重的图的无障碍邻接矩阵的决定因素。最后,提出了一种有效的方法来计算截断的校正因子,包括直到指定长度的所有后向跟踪轨道。 |
课程简介: | We present a new view of Gaussian belief propagation (GaBP) based on a representa- tion of the determinant as a product over or- bits of a graph. We show that the GaBP determinant estimate captures totally back- tracking orbits of the graph and consider how to correct this estimate. We show that the missing orbits may be grouped into equiva- lence classes corresponding to backtrackless orbits and the contribution of each equiv- alence class is easily determined from the GaBP solution. Furthermore, we demon- strate that this multiplicative correction fac- tor can be interpreted as the determinant of a backtrackless adjacency matrix of the graph with edge weights based on GaBP. Finally, an efficient method is proposed to compute a truncated correction factor including all backtrackless orbits up to a specified length. |
关 键 词: | 高斯信念传播; 乘法校正因子; 矩阵 |
课程来源: | 视频讲座网 |
最后编审: | 2019-04-23:lxf |
阅读次数: | 151 |