arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.13860stat.MLcs.LGeess.IV
收藏

考虑结构数据信息的加性张量分解

Additive Tensor Decomposition Considering Structural Data Information

Georgia Institute of Technology(佐治亚理工学院);Georgia Tech Manufacturing Institute(佐治亚理工制造研究所)

Shancong Mou, Andi Wang, Chuck Zhang, Jianjun Shi

AI总结:

针对张量数据的结构信息挖掘需求,该研究提出加性张量分解框架及对应ADMM求解算法,经仿真与医学图像案例验证了其通用性与有效性。

摘要:

具有丰富结构信息的张量数据在过程建模、监控与诊断中愈发重要,此处的结构信息指稀疏性、平滑性、低秩性、分段恒常性等结构属性。为从张量数据中挖掘有效信息,我们提出依据各组分的不同结构信息,将张量分解为多个组分的和。本文给出了张量数据中结构信息的新定义,基于此提出加性张量分解(additive tensor decomposition,ATD)框架以从张量数据中提取有用信息。该框架构建了一个高维优化问题,以获取具备不同结构信息的组分。我们提出交替方向乘子法(alternating direction method of multipliers,ADMM)算法求解该问题,该算法并行性极强,适配所提优化问题。两个仿真算例与一个医学图像分析的真实案例验证了ATD框架的通用性与有效性。

英文摘要:

Tensor data with rich structural information becomes increasingly important in process modeling, monitoring, and diagnosis. Here structural information is referred to structural properties such as sparsity, smoothness, low-rank, and piecewise constancy. To reveal useful information from tensor data, we propose to decompose the tensor into the summation of multiple components based on different structural information of them. In this paper, we provide a new definition of structural information in tensor data. Based on it, we propose an additive tensor decomposition (ATD) framework to extract useful information from tensor data. This framework specifies a high dimensional optimization problem to obtain the components with distinct structural information. An alternating direction method of multipliers (ADMM) algorithm is proposed to solve it, which is highly parallelable and thus suitable for the proposed optimization problem. Two simulation examples and a real case study in medical image analysis illustrate the versatility and effectiveness of the ATD framework.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑