Statistical Performance of Convex Low-Rank and Sparse Tensor Recovery

Suppose a tensor * ∈ ℝn1x...xnK is low-Tucker-rank and sparse simultaneously. The statistical performance of recovering * from it from its noisy observations is studied mathematically in this paper. A convex optimization problem like Remurs [1] which integrates l1-norm and the tensor nuclear norm is proposed. Theoretically, the deterministic upper bound of the estimation error is provided for general noise based on the assumption of restricted strong convexity. For the tensor de-noising problem and the tensor compressive sensing problem, non-asymptotic upper bounds of the estimation error are also shown when the noise is i.i.d. Gaussian.

[1]  Andrzej Cichocki,et al.  Tensor Decompositions for Signal Processing Applications: From two-way to multiway component analysis , 2014, IEEE Signal Processing Magazine.

[2]  Hisashi Kashima,et al.  Statistical Performance of Convex Tensor Decomposition , 2011, NIPS.

[3]  Martin J. Wainwright,et al.  A unified framework for high-dimensional analysis of $M$-estimators with decomposable regularizers , 2009, NIPS.

[4]  Wei Liu,et al.  Tensor Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Tensors via Convex Optimization , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Yonina C. Eldar,et al.  Simultaneously Structured Models With Application to Sparse and Low-Rank Matrices , 2012, IEEE Transactions on Information Theory.

[6]  Haiping Lu,et al.  Multilinear Regression for Embedded Feature Selection with Application to fMRI Analysis , 2017, AAAI.

[7]  Bo Huang,et al.  Square Deal: Lower Bounds and Improved Relaxations for Tensor Recovery , 2013, ICML.

[8]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[9]  Taiji Suzuki,et al.  Convex Tensor Decomposition via Structured Schatten Norm Regularization , 2013, NIPS.

[10]  Tianwen Wei,et al.  Sensing Tensors With Gaussian Filters , 2015, IEEE Transactions on Information Theory.

[11]  Misha Elena Kilmer,et al.  Novel Methods for Multilinear Data Completion and De-noising Based on Tensor-SVD , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Martin J. Wainwright,et al.  Estimation of (near) low-rank matrices with noise and high-dimensional scaling , 2009, ICML.

[13]  Nikos D. Sidiropoulos,et al.  Tensor Decomposition for Signal Processing and Machine Learning , 2016, IEEE Transactions on Signal Processing.

[14]  Jieping Ye,et al.  Tensor Completion for Estimating Missing Values in Visual Data , 2013, IEEE Trans. Pattern Anal. Mach. Intell..