Latent Schatten TT Norm for Tensor Completion

Andong Wang; Xulin Song; Xiyin Wu; Zhihui Lai; Zhong Jin

ICASSP 2019accepted0 citations

Latent Schatten TT Norm for Tensor Completion

Andong Wang, Xulin Song, Xiyin Wu, Zhihui Lai, Zhong Jin

Abstract

Tensor completion arouses much attention in signal processing and machine learning. The tensor train (TT) decomposition has shown better performances than the Tucker decomposition in image and video inpainting. In this paper, we propose a novel tensor completion model based on a newly defined latent Schatten TT norm. Then, the statistical performance is analyzed by establishing a non-asymptotic upper bound on the estimation error. Further, a scalable algorithm is developed to efficiently solve the model. Experimental results of color image inpainting demonstrate that the proposed norm has promising performances compared to other variants of Schatten norm.

BibTeX

@inproceedings{icassp2019_latentschattentt,
  title = {Latent Schatten TT Norm for Tensor Completion},
  author = {Andong Wang and Xulin Song and Xiyin Wu and Zhihui Lai and Zhong Jin},
  booktitle = {ICASSP 2019},
  year = {2019}
}