Big data streaming analytics has recently attracted much attention in the signal and information processing  communities due to the fact that massive streaming datasets have been collected over the years. Among them, many modern data streams are represented as multidimensional arrays (aka tensors), and thus, streaming tensor decomposition or tensor tracking has become a promising tool to analyze such streaming data. In this paper, we propose a novel online algorithm called ROTDL for the problem of robust tensor tracking under the Tucker format. ROTDL is not only capable of tracking the underlying Tucker dictionary of multidimensional data streams over time, but also robust to sparse outliers. The proposed algorithm is specifically designed by using the alternating direction method of multipliers, block-coordinate descent, and recursive least-squares filtering techniques. Several experiments demonstrate the effectiveness of ROTDL for robust tensor tracking.

Abstract: Block-term decomposition (BTD), which factorizes a tensor (aka a multiway array) into block components of low rank, has been a powerful processing tool for multivariate and high-dimensional data analysis. In this paper, we propose a novel tensor tracking method called SBTD for factorizing tensors derived from multidimensional data streams under the BTD format. Thanks to the alternating optimization framework, SBTD first applies a regularized least-squares solver to estimate the temporal factor of the underlying streaming tensor. Then, SBTD adopts an adaptive filter to track the non-temporal tensor factors over time by minimizing a weighted least-squares cost function. Numerical experiments indicate that SBTD is capable of tensor tracking with competitive performance compared to the state-of-the-art BTD algorithms. Accepted at IEEE SSP 2023.

Abstract:  In this paper, we propose two new provable algorithms for tracking online low-rank approximations of high-order streaming tensors with missing data. The first algorithm, dubbed adaptive Tucker decomposition (ATD), minimizes a weighted recursive least-squares cost function to obtain the tensor factors and the core tensor in an efficient way, thanks to the alternating minimization framework and the randomized sketching technique. Under the Canonical Polyadic (CP) model, the second algorithm called ACP is developed as a variant of ATD when the core tensor is imposed to be identity. Both algorithms are low-complexity tensor trackers that have fast convergence and low memory storage requirements. A unified convergence analysis is presented for ATD and ACP to justify their performance. Experiments indicate that the two proposed algorithms are capable of streaming tensor decomposition with competitive performance with respect to estimation accuracy and runtime on both synthetic and real data. Code: https://github.com/thanhtbt/tensor_tracking Comment: to appear in Elsevier Patterns

Abstract: Tensor tracking which is referred to as online (adaptive) decomposition of streaming tensors has recently gained much attention in the signal processing community due to the fact that many modern applications generate a huge number of multidimensional data streams over time. In this paper, we propose an effective tensor tracking method via the tensor-train format for decomposing high-order incomplete streaming tensors. On the arrival of new data, the proposed algorithm minimizes a weighted least-squares objective function accounting for both missing values and time-variation constraints on the underlying tensor-train cores, thanks to the recursive least-squares filtering technique and the block coordinate descent framework. Our algorithm is fully capable of tensor tracking from noisy, incomplete, and high-dimensional observations in both static and time-varying environments. Its tracking ability is validated with several experiments on both synthetic and real data. Code: https://github.com/thanhtbt/ATT-miss/ Comment: Accepted by Signal Processing Journal Â

Tensor decomposition has been demonstrated to be successful in a wide range of applications, from neuroscience and wireless communications to social networks. In an online setting, factorizing tensors derived from multidimensional data streams is however non-trivial due to several inherent problems of real-time stream processing. In recent years, many research efforts have been dedicated to developing online techniques for decomposing such tensors, resulting in significant advances in streaming tensor decomposition or tensor tracking. This topic is emerging and enriches the literature on tensor decomposition, particularly from the data stream analystics perspective. Thus, it is imperative to carry out an overview of tensor tracking to help researchers and practitioners understand its development and achievements, summarise the current trends and advances, and identify challenging problems. In this article, we provide a contemporary and comprehensive survey on different types of tensor tracking techniques. We particularly categorize the state-of-the-art methods into three main groups: streaming CP decompositions, streaming Tucker decompositions, and streaming decompositions under other tensor formats (i.e., tensor-train, t-SVD, and BTD). In each group, we further divide the existing algorithms into sub-categories based on their main optimization framework and model architectures. Finally, we present several research challenges, open problems, and potential directions of tensor tracking in the future.

Abstract: We studied the problem of robust subspace tracking (RST) in contaminated environments. Leveraging the fast approximated power iteration and α-divergence, a novel robust algorithm called αFAPI was developed for tracking the underlying principal subspace of streaming data over time. αFAPI is fast and it outperforms many RST methods while only having a low complexity linear to the data dimension. Some experiments were conducted to illustrate the performance of αFAPI. Matlab Code: https://github.com/thanhtbt/aFAPI Comment: Accepted at IEEE ICASSP 2023