GTT: Leveraging data characteristics for guiding the tensor train decomposition

作者：

Highlights：

• We identify significant relationships among various data characteristics and the accuracies of different tensor train decomposition orders.

• We propose four order selection strategies, (a) aggregate mutual information (AMI), (b) path mutual information (PMI), (c) inverse entropy (IE), and (d) number of parameters (NP), for tensor train decomposition.

• We show that good tensor train orders can be selected through a hybrid (HYB) strategy that takes into account multiple characteristics of the 15 given categorical-valued data set and 3 given continuous-valued data set.

摘要

•We identify significant relationships among various data characteristics and the accuracies of different tensor train decomposition orders.•We propose four order selection strategies, (a) aggregate mutual information (AMI), (b) path mutual information (PMI), (c) inverse entropy (IE), and (d) number of parameters (NP), for tensor train decomposition.•We show that good tensor train orders can be selected through a hybrid (HYB) strategy that takes into account multiple characteristics of the 15 given categorical-valued data set and 3 given continuous-valued data set.

论文关键词：Low-rank embedding,Tensor train decomposition,Order selection

论文评审过程：Received 14 February 2021, Revised 16 March 2022, Accepted 1 April 2022, Available online 4 April 2022, Version of Record 12 May 2022.

论文官网地址：https://doi.org/10.1016/j.is.2022.102047