In this era of big data, data analytics and machine learning, it is imperative to find ways to compress large data sets such that intrinsic features necessary for subsequent analysis are not lost. The traditional workhorse for data dimensionality reduction and feature extraction has been the matrix SVD, which presupposes that the data has been arranged in matrix format. Our main goal in this study is to show that high-dimensional data sets are more compressible when treated as tensors (aka multiway arrays) and compressed via tensor-SVDs under the tensor-tensor product structures in (Kilmer and Martin, 2011; Kernfeld et al., 2015). We begin by proving Eckart Young optimality results for families of tensor-SVDs under two different truncation strategies. As such optimality properties can be proven in both matrix and tensor-based algebras, a fundamental question arises: does the tensor construct subsume the matrix construct in terms of representation efficiency? The answer is yes, as shown when we prove that a tensor-tensor representation of an equal dimensional spanning space can be superior to its matrix counterpart. We then investigate how the compressed representation provided by the truncated tensor-SVD is related both theoretically and in compression performance to its closest tensor-based analogue, truncated HOSVD (De Lathauwer et al., 2000; De Lathauwer and Vandewalle, 2004), thereby showing the potential advantages of our tensor-based algorithms. Finally, we propose new tensor truncated SVD variants, namely multi-way tensor SVDs, provide further approximated representation efficiency and discuss under which conditions they are considered optimal. We conclude with a numerical study demonstrating the utility of the theory.
We propose a framework for discrete scientific data compression based on the tensor-train (TT) decomposition. Our approach is tailored to handle unstructured output data from discrete element method (DEM) simulations, demonstrating its effectiveness in compressing both raw (e.g. particle position and velocity) and derived (e.g. stress and strain) datasets. We show that geometry-driven "tensorization" coupled with the TT decomposition (known as quantized TT) yields a hierarchical compression scheme, achieving high compression ratios for key variables in these DEM datasets.
We present a method for converting tensors into tensor train format based on actions of the tensor as a vector-valued multilinear function. Existing methods for constructing tensor trains require access to "array entries" of the tensor and are therefore inefficient or computationally prohibitive if the tensor is accessible only through its action, especially for high order tensors. Our method permits efficient tensor train compression of large high order derivative tensors for nonlinear mappings that are implicitly defined through the solution of a system of equations. Array entries of these d…
We describe a simple, black-box compression format for tensors with a multiscale structure. By representing the tensor as a sum of compressed tensors defined on increasingly coarse grids, we capture low-rank structures on each grid-scale, and we show how this leads to an increase in compression for a fixed accuracy. We devise an alternating algorithm to represent a given tensor in the multiresolution format and prove local convergence guarantees. In two dimensions, we provide examples that show that this approach can beat the Eckart-Young theorem, and for dimensions higher than two, we achieve…
Tensor decomposition methodologies are proposed to reduce the memory requirement of translation operator tensors arising in the fast multipole method-fast Fourier transform (FMM-FFT)-accelerated surface integral equation (SIE) simulators. These methodologies leverage Tucker, hierarchical Tucker (H-Tucker), and tensor train (TT) decompositions to compress the FFT'ed translation operator tensors stored in three-dimensional (3D) and four-dimensional (4D) array formats. Extensive numerical tests are performed to demonstrate the memory saving achieved by and computational overhead introduced by the…
In recent work (Soltani, Kilmer, Hansen, BIT 2016), an algorithm for non-negative tensor patch dictionary learning in the context of X-ray CT imaging and based on a tensor-tensor product called the $t$-product (Kilmer and Martin, 2011) was presented. Building on that work, in this paper, we use of non-negative tensor patch-based dictionaries trained on other data, such as facial image data, for the purposes of either compression or image deblurring. We begin with an analysis in which we address issues such as suitability of the tensor-based approach relative to a matrix-based approach, diction…
This paper proposes low tensor-train (TT) rank and low multilinear (ML) rank approximations for de-speckling and compression of 3D optical coherence tomography (OCT) images for a given compression ratio (CR). To this end, we derive the alternating direction method of multipliers based algorithms for the related problems constrained with the low TT- and low ML rank. Rank constraints are implemented through the Schatten-p (Sp) norm, p e {0, 1/2, 2/3, 1}, of unfolded matrices. We provide the proofs of global convergence towards a stationary point for both algorithms. Rank adjusted 3D OCT image te…
The need to know a few singular triplets associated with the largest singular values of third-order tensors arises in data compression and extraction. This paper describes a new method for their computation using the t-product. Methods for determining a couple of singular triplets associated with the smallest singular values also are presented. The proposed methods generalize available restarted Lanczos bidiagonalization methods for computing a few of the largest or smallest singular triplets of a matrix. The methods of this paper use Ritz and harmonic Ritz lateral slices to determine accurate…
In this paper, we define a semi-tensor product for third-order tensors. Based on this definition, we present a new type of tensor decomposition strategy and give the specific algorithm. This decomposition strategy actually generalizes the tensor SVD based on semi-tensor product. Due to the characteristic of semi-tensor product for compressing the data scale, we can therefore achieve data compression in this way. Numerical comparisons are given to show the advantages of this decomposition.
Guangdong University of Technology（广东工业大学）；RIKEN（理化学研究所）
Efficient and accurate low-rank approximation (LRA) methods are of great significance for large-scale data analysis. Randomized tensor decompositions have emerged as powerful tools to meet this need, but most existing methods perform poorly in the presence of noise interference. Inspired by the remarkable performance of randomized block Krylov iteration (rBKI) in reducing the effect of tail singular values, this work designs an rBKI-based Tucker decomposition (rBKI-TK) for accurate approximation, together with a hierarchical tensor ring decomposition based on rBKI-TK for efficient compression …
A New Tensor Network: Tubal Tensor Train and Its Applications
Lab of Machine Learning and Knowledge Representation, Innopolis University（机器学习与知识表示实验室，因诺波利斯大学）；Innopolis University（因诺波利斯大学）；Skolkovo Institute of Science and Technology, Center for Artificial Intelligence Technology（斯克尔克沃科学与技术学院，人工智能技术中心）；Systems Research Institute of Polish Academy of Science（波兰科学院系统研究 institute）
