Data prefetching is important for storage system optimization and access performance improvement. Traditional prefetchers work well for mining access patterns of sequential logical block address (LBA) but cannot handle complex non-sequential patterns that commonly exist in real-world applications. The state-of-the-art (SOTA) learning-based prefetchers cover more LBA accesses. However, they do not adequately consider the spatial interdependencies between LBA deltas, which leads to limited performance and robustness. This paper proposes a novel Stream-Graph neural network-based Data Prefetcher (SGDP). Specifically, SGDP models LBA delta streams using a weighted directed graph structure to represent interactive relations among LBA deltas and further extracts hybrid features by graph neural networks for data prefetching. We conduct extensive experiments on eight real-world datasets. Empirical results verify that SGDP outperforms the SOTA methods in terms of the hit ratio by 6.21%, the effective prefetching ratio by 7.00%, and speeds up inference time by 3.13× on average. Besides, we generalize SGDP to different variants by different stream constructions, further expanding its application scenarios and demonstrating its robustness. SGDP offers a novel data prefetching solution and has been verified in commercial hybrid storage systems in the experimental phase. Our codes and appendix are available at https://github.com/yyysjz1997/SGDP/.

Cache plays an important role in storage systems. With better allocation of cache space to each storage device, total I/O latency can be reduced remarkably. To achieve this goal, we propose an Accurate Probabilistic miss ratio curve approximation for Adaptive Cache allocation (APAC) system. APAC can obtain near-optimal performance for allocating cache space with low overhead. Specifically, with a linear-time probabilistic approximation of reuse distance of all blocks inside each device, APAC can accurately estimate the miss ratio curve (MRC). Furthermore, APAC utilizes the MRCs to obtain the near-optimal configuration of cache allocation by dynamic programming. Experimental results show that APAC achieves higher accuracy in MRC approximation compared to the state-of-the-art methods, leading to higher hit ratio and lower latency of the block storage systems.

Tensor singular value decomposition (t-SVD) has recently become increasingly popular for tensor recovery under partial and/or corrupted observations. However, the existing t -SVD-based methods neither make use of a rank prior nor provide an accurate rank estimation (RE), which would limit their recovery performance. From the practical perspective, the tensor RE problem is nontrivial and difficult to solve. In this article, we, therefore, aim to determine the correct rank of an intrinsic low-rank tensor from corrupted observations based on t-SVD and further improve recovery results with the estimated rank. Specifically, we first induce the equivalence of the tensor nuclear norm (TNN) of a tensor and its f -diagonal tensor. We then simultaneously minimize the reconstruction error and TNN of the f -diagonal tensor, leading to RE. Subsequently, we relax our model by removing the TNN regularizer to improve the recovery performance. Furthermore, we consider more general cases in the presence of missing data and/or gross corruptions by proposing robust tensor principal component analysis and robust tensor completion with RE. The robust methods can achieve successful recovery by refining the models with correct estimated ranks. Experimental results show that the proposed methods outperform the state-of-the-art methods with significant improvements.

The discovery and prediction of block access patterns in hybrid storage systems is of crucial importance for effective tier management. Existing methods are usually based on heuristics and unable to handle complex patterns. This work newly introduces transformer to block access pattern prediction. We remark that block accesses in the tier management systems are aggregated temporally and spatially as multivariate time series of block access frequency, so the runtime requirements are relaxed, making complex models applicable for the deployment. Moreover, enormous and rarely accessed blocks in storage systems and the structure of traditional transformer models would result in millions of redundant parameters and make them impractical to be deployed. We incorporate Tensor-Train Decomposition (TTD) with transformer and propose the Compressed Full Tenor Transformer (CFTT), in which all linear layers in the vanilla transformer are replaced with tensor-train layers. Weights of input and output layers are shared to further reduce parameters and reuse knowledge implicitly. CFTT can significantly reduce the model size and computation cost, which is critical to save storage space and inference time. Extensive experiments are conducted on synthetic and real-world datasets. The results demonstrate that transformers achieve state-of-the-art performance stably in terms of top-k hit rates. Moreover, the proposed CFTT compresses transformers 16× to 461× and speeds up inference 5× without sacrificing performance on the whole, which facilitates its applications in tier management in hybrid storage systems.

This article focuses on a new and practical problem in single-sample per person face recognition (SSPP FR), i.e., SSPP FR with a contaminated biometric enrolment database (SSPP-ce FR), where the SSPP-based enrolment database is contaminated by nuisance facial variations in the wild, such as poor lightings, expression change, and disguises (e.g., wearing sunglasses, hat, and scarf). In SSPP-ce FR, the most popular generic learning methods will suffer serious performance degradation because the prototype plus variation (P+V) model used in these methods is no longer suitable in such scenarios. The reasons are twofold. First, the contaminated enrolment samples could yield bad prototypes to represent the persons. Second, the generated variation dictionary is simply based on the subtraction of the average face from generic samples of the same person and cannot well depict the intrapersonal variations. To address the SSPP-ce FR problem, we propose a novel iterative dynamic generic learning (IDGL) method, where the labeled enrolment database and the unlabeled query set are fed into a dynamic label feedback network for learning. Specifically, IDGL first recovers the prototypes for the contaminated enrolment samples via a semisupervised low-rank representation (SSLRR) framework and learns a representative variation dictionary by extracting the "sample-specific" corruptions from an auxiliary generic set. Then, it puts them into the P+V model to estimate labels for query samples. Subsequently, the estimated labels will be used as feedback to modify the SSLRR, thus updating new prototypes for the next round of P+V-based label estimation. With the dynamic learning network, the accuracy of the estimated labels is improved iteratively by virtue of the steadily enhanced prototypes. Experiments on various benchmark face data sets have demonstrated the superiority of IDGL over state-of-the-art counterparts.

This work proposes a novel approach for multiple time series forecasting. At first, multi-way delay embedding transform (MDT) is employed to represent time series as low-rank block Hankel tensors (BHT). Then, the higher-order tensors are projected to compressed core tensors by applying Tucker decomposition. At the same time, the generalized tensor Autoregressive Integrated Moving Average (ARIMA) is explicitly used on consecutive core tensors to predict future samples. In this manner, the proposed approach tactically incorporates the unique advantages of MDT tensorization (to exploit mutual correlations) and tensor ARIMA coupled with low-rank Tucker decomposition into a unified framework. This framework exploits the low-rank structure of block Hankel tensors in the embedded space and captures the intrinsic correlations among multiple TS, which thus can improve the forecasting results, especially for multiple short time series. Experiments conducted on three public datasets and two industrial datasets verify that the proposed BHT-ARIMA effectively improves forecasting accuracy and reduces computational cost compared with the state-of-the-art methods.

This paper studies a new challenging problem in face recognition (FR) with single sample per person (SSPP), i.e., SSPP FR with a contaminated gallery (SSPP-CG FR), where the gallery is contaminated by variations. In SSPP-CG FR, the popular generic learning methods will suffer serious performance degradation because the applied prototype plus variation (P+V) model is not suitable in such scenarios. The reasons are twofold: 1) The contaminated gallery samples yield bad prototypes to represent the persons; 2) The generated variation dictionary is simply based on the subtraction of average face from generic samples of the same person and cannot well depict the intra-personal variations. To tackle SSPPCG FR, we propose a novel Iterative Dynamic Generic Learning (IDGL) method, where the labeled gallery and unlabeled query sets are fed into a dynamic label feedback network for learning. Specifically, IDGL first recovers the prototypes via a semi-supervised low-rank representation (SSLRR) framework and learns a representative variation dictionary by extracting the “sample-specific” corruptions from an auxiliary generic set. Then, it puts them into the P+V model to estimate labels for query samples. Subsequently, the estimated labels are used as the feedbacks to modify the SSLRR, thus updating new prototypes for the next round of P+V based label estimation. With the dynamic learning network, the accuracy of the estimated labels is improved iteratively owing to the steadily enhanced prototypes. Experiments on various benchmark databases have verified the superiority of IDGL.

In this paper, the problem of time series reconstruction in a multiway delay embedded space using Tensor Train decomposition is addressed. A new algorithm has been developed in which an incomplete signal is first transformed to a Hankel matrix and in the next step to a higher order tensor using extended Multiway Delay embedded Transform. Then, the resulting higher order tensor is completed using low rank Tensor Train decomposition. Comparing to previous Hankelization approaches, in the proposed approach, blocks of elements are used for Hankelization instead of individual elements, which results in producing a higher order tensor. Simulation results confirm the effectiveness and high performance of the proposed completion approach. Although in this paper we focus on single time series, our method can be straightforwardly extended to reconstruction of multivariate time series, color images and videos.

Multidimensional data (i.e., tensors) with missing entries are common in practice. Extracting features from incomplete tensors is an important yet challenging problem in many fields such as machine learning, pattern recognition, and computer vision. Although the missing entries can be recovered by tensor completion techniques, these completion methods focus only on missing data estimation instead of effective feature extraction. To the best of our knowledge, the problem of feature extraction from incomplete tensors has yet to be well explored in the literature. In this paper, we therefore tackle this problem within the unsupervised learning environment. Specifically, we incorporate low-rank tensor decomposition with feature variance maximization (TDVM) in a unified framework. Based on orthogonal Tucker and CP decompositions, we design two TDVM methods, TDVM-Tucker and TDVM-CP, to learn low-dimensional features viewing the core tensors of the Tucker model as features and viewing the weight vectors of the CP model as features. TDVM explores the relationship among data samples via maximizing feature variance and simultaneously estimates the missing entries via low-rank Tucker/CP approximation, leading to informative features extracted directly from observed entries. Furthermore, we generalize the proposed methods by formulating a general model that incorporates feature regularization into low-rank tensor approximation. In addition, we develop a joint optimization scheme to solve the proposed methods by integrating the alternating direction method of multipliers with the block coordinate descent method. Finally, we evaluate our methods on six real-world image and video data sets under a newly designed multiblock missing setting. The extracted features are evaluated in face recognition, object/action classification, and face/gait clustering. Experimental results demonstrate the superior performance of the proposed methods compared with the state-of-the-art approaches.

Completing a matrix from a small subset of its entries, i.e., matrix completion is a challenging problem arising from many real-world applications, such as machine learning and computer vision. One popular approach to solve the matrix completion problem is based on low-rank decomposition/factorization. Low-rank matrix decomposition-based methods often require a prespecified rank, which is difficult to determine in practice. In this paper, we propose a novel low-rank decomposition-based matrix completion method with automatic rank estimation. Our method is based on rank-one approximation, where a matrix is represented as a weighted summation of a set of rank-one matrices. To automatically determine the rank of an incomplete matrix, we impose L1-norm regularization on the weight vector and simultaneously minimize the reconstruction error. After obtaining the rank, we further remove the L1-norm regularizer and refine recovery results. With a correctly estimated rank, we can obtain the optimal solution under certain conditions. Experimental results on both synthetic and real-world data demonstrate that the proposed method not only has good performance in rank estimation, but also achieves better recovery accuracy than competing methods.

Tensor completion (TC) is a challenging problem of recovering missing entries of a tensor from its partial observation. One main TC approach is based on CP/Tucker decomposition. However, this approach often requires the determination of a tensor rank a priori. This rank estimation problem is difficult in practice. Several Bayesian solutions have been proposed but they often under/over-estimate the tensor rank while being quite slow. To address this problem of rank estimation with missing entries, we view the weight vector of the orthogonal CP decomposition of a tensor to be analogous to the vector of singular values of a matrix. Subsequently, we define a new CP-based tensor nuclear norm as the $L_1$-norm of this weight vector. We then propose Tensor Rank Estimation based on $L_1$-regularized orthogonal CP decomposition (TREL1) for both CP-rank and Tucker-rank. Specifically, we incorporate a regularization with CP-based tensor nuclear norm when minimizing the reconstruction error in TC to automatically determine the rank of an incomplete tensor. Experimental results on both synthetic and real data show that: 1) Given sufficient observed entries, TREL1 can estimate the true rank (both CP-rank and Tucker-rank) of incomplete tensors well; 2) The rank estimated by TREL1 can consistently improve recovery accuracy of decomposition-based TC methods; 3) TREL1 is not sensitive to its parameters in general and more efficient than existing rank estimation methods.