Weighted Tensor Decomposition for Learning Latent Variables with Partial Data
Weighted Tensor Decomposition for Learning Latent Variables with Partial Data
复制标题
使用部分数据学习潜变量的加权张量分解
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
F. Doshi
中科院分区：
文献类型：
--
作者：
Omer Gottesman;Weiwei Pan;F. Doshi
张量分解方法是用于仅给定数据的低阶矩来学习潜变量的流行工具。然而，标准的假设是我们有足够的数据来高精度地估计这些时刻。在这项工作中，我们考虑的情况下，在某些方面的数据并不总是观察-常见的应用设置，其中不是所有的测量可能采取的所有观测-导致不同质量的时刻估计。我们推导出一种加权张量分解方法，该方法在计算上与非加权方法一样有效，并证明它优于不适当利用这些较少观察到的维度的方法。
Tensor decomposition methods are popular tools for learning latent variables given only lower-order moments of the data. However, the standard assumption is that we have sufficient data to estimate these moments to high accuracy. In this work, we consider the case in which certain dimensions of the data are not always observed---common in applied settings, where not all measurements may be taken for all observations---resulting in moment estimates of varying quality. We derive a weighted tensor decomposition approach that is computationally as efficient as the non-weighted approach, and demonstrate that it outperforms methods that do not appropriately leverage these less-observed dimensions.