Distinguishing key features in complex visual tasks is challenging. A novel approach treats image patches (tokens) as waves. By using both phase and amplitude, it captures richer semantics and specific invariances compared to pixel-based methods, and allows for feature fusion across regions for a holistic image representation. Based on this, we propose the Wave-wise Discriminative Transformer Tracker (WDT). During tracking, WDT represents features via phase-amplitude separation, enhancement, and mixture. First, we designed a Mutual Exclusive Phase-Amplitude Extractor (MEPAE) to separate phase and amplitude features with distinct semantics, representing spatial target info and background brightness respectively. Then, Wave-wise Feature Augmentation is carried out with two submodules: Phase-Amplitude Feature Augmentation and Mixture. The augmentation module disrupts the separated features in the same batch, and the mixture module recombines them to generate positive and negative waves. The original features are aggregated into the original wave. Positive waves have the same phase but different amplitudes, and negative waves have different phase components. Finally, self-supervised and tracking-supervised losses guide the global and local representation learning for original, positive, and negative waves, enhancing wave-level discrimination. Experiments on five benchmarks prove the effectiveness of our method.

Multi-view clustering (MVC) is a fast-growing research direction. However, most existing MVC works focus on concrete objects (e.g., cats, desks) but ignore abstract objects (e.g., knowledge, thoughts), which are also important parts of our daily lives and more correlated to cognition. Relational knowledge, as a typical abstract concept, describes the relationship between entities. For example, "Cats like eating fishes," as relational knowledge, reveals the relationship "eating" between "cats" and "fishes." To fill this gap, we first point out that MVC on relational knowledge is considered an important scenario. Then, we construct 8 new datasets to lay research grounds for them. Moreover, a simple yet effective relational knowledge MVC paradigm (RK-MVC) is proposed by compensating the omitted sample-global correlations from the structural knowledge information. Concretely, the basic consensus features are first learned via adopted MVC backbones, and sample-global correlations are generated in both coarse-grained and fine-grained manners. In particular, the sample-global correlation learning module can be easily extended to various MVC backbones. Finally, both basic consensus features and sample-global correlation features are weighted fused as the target consensus feature. We adopt 9 typical MVC backbones in this paper for comparison from 7 aspects, demonstrating the promising capacity of our RK-MVC.

Site selection aims to select optimal locations for new stores, which is crucial in business management and urban computing. The early data-driven models heavily relied on feature engineering, which could not effectively model the complex relationships and diverse influences among different data. To alleviate such issues, the knowledge-driven paradigm is proposed based on urban knowledge graphs (KGs). However, the research on them is at an early stage. They omit extra multi-modal information corresponding to brands and stores due to two main challenges, i.e., (1) building available datasets, and (2) designing effective models. It constrains the expressive ability and practical value of previous models. To this end, we first construct new multi-modal urban KGs for site selection with three extra modal (i.e., visual, textual, and acoustic) attributes. Then, we propose a novel multi-modal knowledge-driven model (MGKsite). Concretely, a graph neural network (GNN) based fusion network is designed to fuse the features based on the attribute K-Nearest Neighbor (KNN) graph, which models both intra and inter-modal correlations among the features. The fused embeddings are further injected into the knowledge-driven backbones for learning and inference. Experiments prove promising capacities of MGKsite from five aspects, i.e., superiority, effectiveness, sensitivity, transferability and complexity.

Traditional knowledge graph embedding (KGE) models map entities and relations to unique embedding vectors in a shallow lookup manner. As the scale of data becomes larger, this manner will raise unaffordable computational costs. Anchor-based strategies have been treated as effective ways to alleviate such efficiency problems by propagation on representative entities instead of the whole graph. However, most existing anchor-based KGE models select the anchors in a primitive manner, which limits their performance. To this end, we propose a novel anchor-based strategy for KGE, i.e., a relational clustering-based anchor selection strategy (RecPiece), where two characteristics are leveraged, i.e., (1) representative ability of the cluster centroids and (2) descriptive ability of relation types in KGs. Specifically, we first perform clustering over features of factual triplets instead of entities, where cluster number is naturally set as number of relation types since each fact can be characterized by its relation in KGs. Then, representative triplets are selected around the clustering centroids, further mapped into corresponding anchor entities. Extensive experiments on six datasets show that RecPiece achieves higher performances but comparable or even fewer parameters compared to previous anchor-based KGE models, indicating that our model can select better anchors in a more scalable way.

Generalized sparse nonnegative matrix factorization (SNMF) has been proven useful in extracting information and representing sparse data with various types of probabilistic distributions from industrial applications, e.g., recommender systems and social networks.However, current solution approaches for generalized SNMF are based on the manipulation of whole sparse matrices and factor matrices, which will result in large-scale intermediate data.Thus, these approaches cannot describe the high-dimensional and sparse matrices in mainstream industrial and big data platforms, e.g., graphics processing unit (GPU) and multi-GPU, in an online and scalable manner. To overcome these issues, an online, scalable, and single-thread-based SNMF for CUDA parallelization on GPU (CUSNMF) and multi-GPU (MCUSNMF) is proposed in this article. First, theoretical derivation is conducted, which demonstrates that the CUSNMF depends only on the products and sums of the involved feature tuples. Next, the compactness, which can follow the sparsity pattern of sparse matrices, endows the CUSNMF with online learning capability and the fine granularity gives it high parallelization potential on GPU and multi-GPU. Finally, the performance results on several real industrial datasets demonstrate the linear scalability of the time overhead and the space requirement and the validity of the extension to online learning. Moreover, CUSNMF obtains speedup of 7X on a P100 GPU compared to that of the state-of-the-art parallel approaches on a shared memory platform.

Due to high capacity and fast transmission speed, 5G plays a key role in modern electronic infrastructure. Meanwhile, sparse tensor factorization (STF) is a useful tool for dimension reduction to analyze high-order, high-dimension, and sparse tensor (HOHDST) data, which is transmitted on 5G Internet-of-things (IoT). Hence, HOHDST data relies on STF to obtain complete data and discover rules for real time and accurate analysis. From another view of computation and data security, the current STF solution seeks to improve the computational efficiency but neglects privacy security of the IoT data, e.g., data analysis for network traffic monitor system. To overcome these problems, this article proposes a multiple-strategies differential privacy framework on STF (MDPSTF) for HOHDST network traffic data analysis.MDPSTFcomprises three differential privacy (DP) mechanisms, i.e.,$\varepsilon -$DP, concentrated DP, and local DP. Furthermore, the theoretical proof of privacy bound is presented. Hence,MDPSTFcan provide general data protection for HOHDST network traffic data with high-security promise. We conduct experiments on two real network traffic datasets ($Abilene$and$G\grave{E}ANT$). The experimental results show thatMDPSTFhas high universality on the various degrees of privacy protection demands and high recovery accuracy for the HOHDST network traffic data.

Sparse Tucker Decomposition (STD) algorithms learn a core tensor and a group of factor matrices to obtain an optimal low-rank representation feature for the High-Order, High-Dimension, and Sparse Tensor (HOHDST). However, existing STD algorithms face the problem of intermediate variables explosion which results from the fact that the formation of those variables, i.e., matrices Khatri-Rao product, Kronecker product, and matrix-matrix multiplication, follows the whole elements in sparse tensor. The above problems prevent deep fusion of efficient computation and big data platforms. To overcome the bottleneck, a novel stochastic optimization strategy (SGD Tucker) is proposed for STD which can automatically divide the high-dimension intermediate variables into small batches of intermediate matrices. Specifically, SGD Tucker only follows the randomly selected small samples rather than the whole elements, while maintaining the overall accuracy and convergence rate. In practice, SGD Tucker features the two distinct advancements over the state of the art. First, SGD Tucker can prune the communication overhead for the core tensor in distributed settings. Second, the low data-dependence of SGD Tucker enables fine-grained parallelization, which makes SGD Tucker obtaining lower computational overheads with the same accuracy. Experimental results show that SGD Tucker runs at least 2X faster than the state of the art.

Given a high-order, large-scale and sparse data from big data and industrial applications, how can we acquire useful patterns in a real-time and low memory overhead manner? Sparse Non-negative tensor factorization (SNTF) possesses high-order representation, non-negativity and dimension reduction inherence. Thus, SNTF has become a useful tool to represent and analyze the sparse data, which has been incorporated with extra contextual information, i.e., time and location, etc, more than the matrix, which can only model the 2 ways data. However, current SNTF techniques suffer from a) non-linear time and space overhead, b) intermediate data explosion, and c) inability on GPU and multi-GPU. To address these issues, a single-thread-based SNTF is proposed, which involves the feature elements rather than on the whole factor matrices, and can avoid the forming of large-scale intermediate matrices. Then, a CUDA parallelizing single-thread-based SNTF (CUSNTF) model is proposed for industrial applications on GPU and multi-GPU (MCUSNTF). Thus, CUSNTF has linear computing and space complexity, and linear communication cost on multi-GPU. We implement CUSNTF and MCUSNTF on 8 P100 GPUs, and compare it with state-of-the-art parallel and distributed methods. Experimental results from several industrial datasets demonstrate that the linear scalability and efficiency of CUSNTF.

Real-time accurate recommendation of large-scale recommender systems is a challenging task. Matrix factorization (MF), as one of the most accurate and scalable techniques to predict missing ratings, has become popular in the collaborative filtering (CF) community. Currently, stochastic gradient descent (SGD) is one of the most famous approaches for MF. However, it is non-trivial to parallelize SGD for large-scale CF MF problems due to the dependence on the user and item pair, which can cause parallelization over-writing. To remove the dependence on the user and item pair, we propose a multi-stream SGD (MSGD) approach, for which the update process is theoretically convergent. On that basis, we propose a Compute Unified Device Architecture (CUDA) parallelization MSGD (CUMSGD) approach. CUMSGD can obtain high parallelism and scalability on Graphic Processing Units (GPUs). On Tesla K20m and K40c GPUs, the experimental results show that CUMSGD outperforms prior works that accelerated MF on shared memory systems, e.g., DSGD, FPSGD, Hogwild!, and CCD++. For large-scale CF problems, we propose multiple GPUs (multi-GPU) CUMSGD (MCUMSGD). The experimental results show that MCUMSGD can improve MSGD performance further. With a K20m GPU card, CUMSGD can be 5-10 times as fast compared with the state-of-the-art approaches on shared memory platform.