Apr 2023, Volume 17 Issue 2
    

  • Select all
    Architecture
  • RESEARCH ARTICLE
    Huize LI, Hai JIN, Long ZHENG, Yu HUANG, Xiaofei LIAO

    With the increasing amount of data, there is an urgent need for efficient sorting algorithms to process large data sets. Hardware sorting algorithms have attracted much attention because they can take advantage of different hardware’s parallelism. But the traditional hardware sort accelerators suffer “memory wall” problems since their multiple rounds of data transmission between the memory and the processor. In this paper, we utilize the in-situ processing ability of the ReRAM crossbar to design a new ReCAM array that can process the matrix-vector multiplication operation and the vector-scalar comparison in the same array simultaneously. Using this designed ReCAM array, we present ReCSA, which is the first dedicated ReCAM-based sort accelerator. Besides hardware designs, we also develop algorithms to maximize memory utilization and minimize memory exchanges to improve sorting performance. The sorting algorithm in ReCSA can process various data types, such as integer, float, double, and strings.

    We also present experiments to evaluate the performance and energy efficiency against the state-of-the-art sort accelerators. The experimental results show that ReCSA has 90.92×, 46.13×, 27.38×, 84.57×, and 3.36× speedups against CPU-, GPU-, FPGA-, NDP-, and PIM-based platforms when processing numeric data sets. ReCSA also has 24.82×, 32.94×, and 18.22× performance improvement when processing string data sets compared with CPU-, GPU-, and FPGA-based platforms.

  • Software
  • RESEARCH ARTICLE
    Huan ZHOU, Weining QIAN, Xuan ZHOU, Qiwen DONG, Aoying ZHOU, Wenrong TAN

    On-line transaction processing (OLTP) systems rely on transaction logging and quorum-based consensus protocol to guarantee durability, high availability and strong consistency. This makes the log manager a key component of distributed database management systems (DDBMSs). The leader of DDBMSs commonly adopts a centralized logging method to writing log entries into a stable storage device and uses a constant log replication strategy to periodically synchronize its state to followers. With the advent of new hardware and high parallelism of transaction processing, the traditional centralized design of logging limits scalability, and the constant trigger condition of replication can not always maintain optimal performance under dynamic workloads.

    In this paper, we propose a new log manager named Salmo with scalable logging and adaptive replication for distributed database systems. The scalable logging eliminates centralized contention by utilizing a highly concurrent data structure and speedy log hole tracking. The kernel of adaptive replication is an adaptive log shipping method, which dynamically adjusts the number of log entries transmitted between leader and followers based on the real-time workload. We implemented and evaluated Salmo in the open-sourced transaction processing systems Cedar and DBx1000. Experimental results show that Salmo scales well by increasing the number of working threads, improves peak throughput by 1.56× and reduces latency by more than 4× over log replication of Raft, and maintains efficient and stable performance under dynamic workloads all the time.

  • RESEARCH ARTICLE
    Changbo KE, Fu XIAO, Zhiqiu HUANG, Fangxiong XIAO

    In an ever-changing environment, Software as a Service (SaaS) can rarely protect users’ privacy. Being able to manage and control the privacy is therefore an important goal for SaaS. Once the participant of composite service is substituted, it is unclear whether the composite service satisfy user privacy requirement or not. In this paper, we propose a privacy policies automatic update method to enhance user privacy when a service participant change in the composite service. Firstly, we model the privacy policies and service variation rules. Secondly, according to the service variation rules, the privacy policies are automatically generated through the negotiation between user and service composer. Thirdly, we prove the feasibility and applicability of our method with the experiments. When the service quantity is 50, ratio that the services variations are successfully checked by monitor is 81%. Moreover, ratio that the privacy policies are correctly updated is 93.6%.

  • RESEARCH ARTICLE
    Chunxi ZHANG, Yuming LI, Rong ZHANG, Weining QIAN, Aoying ZHOU

    Massive scale of transactions with critical requirements become popular for emerging businesses, especially in E-commerce. One of the most representative applications is the promotional event running on Alibaba’s platform on some special dates, widely expected by global customers. Although we have achieved significant progress in improving the scalability of transactional database systems (OLTP), the presence of contention operations in workloads is still one of the fundamental obstacles to performance improving. The reason is that the overhead of managing conflict transactions with concurrency control mechanisms is proportional to the amount of contentions. As a consequence, generating contented workloads is urgent to evaluate performance of modern OLTP database systems. Though we have kinds of standard benchmarks which provide some ways in simulating contentions, e.g., skew distribution control of transactions, they can not control the generation of contention quantitatively; even worse, the simulation effectiveness of these methods is affected by the scale of data. So in this paper we design a scalable quantitative contention generation method with fine contention granularity control. We conduct a comprehensive set of experiments on popular opensourced DBMSs compared with the latest contention simulation method to demonstrate the effectiveness of our generation work.

  • Artificial Intelligence
  • LETTER
    Song XIAO, Ting BAI, Xiangchong CUI, Bin WU, Xinkai MENG, Bai WANG
  • RESEARCH ARTICLE
    Junxiao XUE, Hao ZHOU

    Biometric speech recognition systems are often subject to various spoofing attacks, the most common of which are speech synthesis and speech conversion attacks. These spoofing attacks can cause the biometric speech recognition system to incorrectly accept these spoofing attacks, which can compromise the security of this system. Researchers have made many efforts to address this problem, and the existing studies have used the physical features of speech to identify spoofing attacks. However, recent studies have shown that speech contains a large number of physiological features related to the human face. For example, we can determine the speaker’s gender, age, mouth shape, and other information by voice. Inspired by the above researches, we propose a spoofing attack recognition method based on physiological-physical features fusion. This method involves feature extraction, a densely connected convolutional neural network with squeeze and excitation block (SE-DenseNet), and feature fusion strategies. We first extract physiological features in audio from a pre-trained convolutional network. Then we use SE-DenseNet to extract physical features. Such a dense connection pattern has high parameter efficiency, and squeeze and excitation blocks can enhance the transmission of the feature. Finally, we integrate the two features into the classification network to identify the spoofing attacks. Experimental results on the ASVspoof 2019 data set show that our model is effective for voice spoofing detection. In the logical access scenario, our model improves the tandem decision cost function and equal error rate scores by 5% and 7%, respectively, compared to existing methods.

  • RESEARCH ARTICLE
    Qingyang LI, Zhiwen YU, Huang XU, Bin GUO

    Anomaly detectors are used to distinguish differences between normal and abnormal data, which are usually implemented by evaluating and ranking the anomaly scores of each instance. A static unsupervised streaming anomaly detector is difficult to dynamically adjust anomaly score calculation. In real scenarios, anomaly detection often needs to be regulated by human feedback, which benefits adjusting anomaly detectors. In this paper, we propose a human-machine interactive streaming anomaly detection method, named ISPForest, which can be adaptively updated online under the guidance of human feedback. In particular, the feedback will be used to adjust the anomaly score calculation and structure of the detector, ideally attaining more accurate anomaly scores in the future. Our main contribution is to improve the tree-based streaming anomaly detection model that can be updated online from perspectives of anomaly score calculation and model structure. Our approach is instantiated for the powerful class of tree-based streaming anomaly detectors, and we conduct experiments on a range of benchmark datasets. The results demonstrate that the utility of incorporating feedback can improve the performance of anomaly detectors with a few human efforts.

  • RESEARCH ARTICLE
    Zhe XUE, Junping DU, Xin XU, Xiangbin LIU, Junfu WANG, Feifei KOU

    Node classification has a wide range of application scenarios such as citation analysis and social network analysis. In many real-world attributed networks, a large portion of classes only contain limited labeled nodes. Most of the existing node classification methods cannot be used for few-shot node classification. To train the model effectively and improve the robustness and reliability of the model with scarce labeled samples, in this paper, we propose a local adaptive discriminant structure learning (LADSL) method for few-shot node classification. LADSL aims to properly represent the nodes in the attributed graphs and learn a metric space with a strong discriminating power by reducing the intra-class variations and enlarging inter-class differences. Extensive experiments conducted on various attributed networks datasets demonstrate that LADSL is superior to the other methods on few-shot node classification task.

  • RESEARCH ARTICLE
    Dantong OUYANG, Mengting LIAO, Yuxin YE

    In description logic, axiom pinpointing is used to explore defects in ontologies and identify hidden justifications for a logical consequence. In recent years, SAT-based axiom pinpointing techniques, which rely on the enumeration of minimal unsatisfiable subsets (MUSes) of pinpointing formulas, have gained increasing attention. Compared with traditional Tableau-based reasoning approaches, SAT-based techniques are more competitive when computing justifications for consequences in large-scale lightweight description logic ontologies. In this article, we propose a novel enumeration justification algorithm, working with a replicated driver. The replicated driver discovers new justifications from the explored justifications through cheap literals resolution, which avoids frequent calls of SAT solver. Moreover, when the use of SAT solver is inevitable, we adjust the strategies and heuristic parameters of the built-in SAT solver of axiom pinpointing algorithm. The adjusted SAT solver is able to improve the checking efficiency of unexplored sub-formulas. Our proposed method is implemented as a tool named RDMinA. The experimental results show that RDMinA outperforms the existing axiom pinpointing tools on practical biomedical ontologies such as Gene, Galen, NCI and Snomed-CT.

  • RESEARCH ARTICLE
    Zhen WU, Xinyu DAI, Rui XIA

    Emotion-cause pair extraction (ECPE) aims to extract all the pairs of emotions and corresponding causes in a document. It generally contains three subtasks, emotions extraction, causes extraction, and causal relations detection between emotions and causes. Existing works adopt pipelined approaches or multi-task learning to address the ECPE task. However, the pipelined approaches easily suffer from error propagation in real-world scenarios. Typical multi-task learning cannot optimize all tasks globally and may lead to suboptimal extraction results. To address these issues, we propose a novel framework, Pairwise Tagging Framework (PTF), tackling the complete emotion-cause pair extraction in one unified tagging task. Unlike prior works, PTF innovatively transforms all subtasks of ECPE, i.e., emotions extraction, causes extraction, and causal relations detection between emotions and causes, into one unified clause-pair tagging task. Through this unified tagging task, we can optimize the ECPE task globally and extract more accurate emotion-cause pairs. To validate the feasibility and effectiveness of PTF, we design an end-to-end PTF-based neural network and conduct experiments on the ECPE benchmark dataset. The experimental results show that our method outperforms pipelined approaches significantly and typical multi-task learning approaches.

  • RESEARCH ARTICLE
    Yi ZHONG, Jia-Hui PAN, Haoxin LI, Wei-Shi ZHENG

    Anticipating future actions without observing any partial videos of future actions plays an important role in action prediction and is also a challenging task. To obtain abundant information for action anticipation, some methods integrate multimodal contexts, including scene object labels. However, extensively labelling each frame in video datasets requires considerable effort. In this paper, we develop a weakly supervised method that integrates global motion and local fine-grained features from current action videos to predict next action label without the need for specific scene context labels. Specifically, we extract diverse types of local features with weakly supervised learning, including object appearance and human pose representations without ground truth. Moreover, we construct a graph convolutional network for exploiting the inherent relationships of humans and objects under present incidents. We evaluate the proposed model on two datasets, the MPII-Cooking dataset and the EPIC-Kitchens dataset, and we demonstrate the generalizability and effectiveness of our approach for action anticipation.

  • RESEARCH ARTICLE
    Zhong JI, Jingwei NI, Xiyao LIU, Yanwei PANG

    Although few-shot learning (FSL) has achieved great progress, it is still an enormous challenge especially when the source and target set are from different domains, which is also known as cross-domain few-shot learning (CD-FSL). Utilizing more source domain data is an effective way to improve the performance of CD-FSL. However, knowledge from different source domains may entangle and confuse with each other, which hurts the performance on the target domain. Therefore, we propose team-knowledge distillation networks (TKD-Net) to tackle this problem, which explores a strategy to help the cooperation of multiple teachers. Specifically, we distill knowledge from the cooperation of teacher networks to a single student network in a meta-learning framework. It incorporates task-oriented knowledge distillation and multiple cooperation among teachers to train an efficient student with better generalization ability on unseen tasks. Moreover, our TKD-Net employs both response-based knowledge and relation-based knowledge to transfer more comprehensive and effective knowledge. Extensive experimental results on four fine-grained datasets have demonstrated the effectiveness and superiority of our proposed TKD-Net approach.

  • RESEARCH ARTICLE
    Xianfeng LIANG, Shuheng SHEN, Enhong CHEN, Jinchang LIU, Qi LIU, Yifei CHENG, Zhen PAN

    Distributed stochastic gradient descent and its variants have been widely adopted in the training of machine learning models, which apply multiple workers in parallel. Among them, local-based algorithms, including LocalSGD and FedAvg, have gained much attention due to their superior properties, such as low communication cost and privacy-preserving. Nevertheless, when the data distribution on workers is non-identical, local-based algorithms would encounter a significant degradation in the convergence rate. In this paper, we propose Variance Reduced Local SGD (VRL-SGD) to deal with the heterogeneous data. Without extra communication cost, VRL-SGD can reduce the gradient variance among workers caused by the heterogeneous data, and thus it prevents local-based algorithms from slow convergence rate. Moreover, we present VRL-SGD-W with an effective warm-up mechanism for the scenarios, where the data among workers are quite diverse. Benefiting from eliminating the impact of such heterogeneous data, we theoretically prove that VRL-SGD achieves a linear iteration speedup with lower communication complexity even if workers access non-identical datasets. We conduct experiments on three machine learning tasks. The experimental results demonstrate that VRL-SGD performs impressively better than Local SGD for the heterogeneous data and VRL-SGD-W is much robust under high data variance among workers.

  • RESEARCH ARTICLE
    Hongsheng XU, Zihan CHEN, Yu ZHANG, Xin GENG, Siya MI, Zhihong YANG

    Temporal localization is crucial for action video recognition. Since the manual annotations are expensive and time-consuming in videos, temporal localization with weak video-level labels is challenging but indispensable. In this paper, we propose a weakly-supervised temporal action localization approach in untrimmed videos. To settle this issue, we train the model based on the proxies of each action class. The proxies are used to measure the distances between action segments and different original action features. We use a proxy-based metric to cluster the same actions together and separate actions from backgrounds. Compared with state-of-the-art methods, our method achieved competitive results on the THUMOS14 and ActivityNet1.2 datasets.

  • Theoretical Computer Science
  • RESEARCH ARTICLE
    Peng YANG, Zhiguo FU

    Local holographic transformations were introduced by Cai et al., and local affine functions, an extra tractable class, were derived by it in #CSP2. In the present paper, we not only generalize local affine functions to #CSPd for general d, but also give new tractable classes by combining local holographic transformations with global holographic transformations. Moreover, we show how to use local holographic transformations to prove hardness. This is of independent interests in the complexity classification of counting problems.

  • Networks and Communication
  • RESEARCH ARTICLE
    Abdelfettah MAATOUG, Ghalem BELALEM, Saïd MAHMOUDI

    Nowadays, smart buildings rely on Internet of things (IoT) technology derived from the cloud and fog computing paradigms to coordinate and collaborate between connected objects. Fog is characterized by low latency with a wider spread and geographically distributed nodes to support mobility, real-time interaction, and location-based services. To provide optimum quality of user life in modern buildings, we rely on a holistic Framework, designed in a way that decreases latency and improves energy saving and services efficiency with different capabilities. Discrete EVent system Specification (DEVS) is a formalism used to describe simulation models in a modular way. In this work, the sub-models of connected objects in the building are accurately and independently designed, and after installing them together, we easily get an integrated model which is subject to the fog computing Framework. Simulation results show that this new approach significantly, improves energy efficiency of buildings and reduces latency. Additionally, with DEVS, we can easily add or remove sub-models to or from the overall model, allowing us to continually improve our designs.

  • Information Systems
  • LETTER
    Ning WANG, Wei ZHENG, Zhigang WANG, Zhiqiang WEI, Yu GU, Peng TANG, Ge YU
  • RESEARCH ARTICLE
    Na LI, Huaijie ZHU, Wenhao LU, Ningning CUI, Wei LIU, Jian YIN, Jianliang XU, Wang-Chien LEE

    Recently a lot of works have been investigating to find the tenuous groups, i.e., groups with few social interactions and weak relationships among members, for reviewer selection and psycho-educational group formation. However, the metrics (e.g., k-triangle, k-line, and k-tenuity) used to measure the tenuity, require a suitable k value to be specified which is difficult for users without background knowledge. Thus, in this paper we formulate the most tenuous group (MTG) query in terms of the group distance and average group distance of a group measuring the tenuity to eliminate the influence of parameter k on the tenuity of the group. To address the MTG problem, we first propose an exact algorithm, namely MTG-VDIS, which takes priority to selecting those vertices whose vertex distance is large, to generate the result group, and also utilizes effective filtering and pruning strategies. Since MTG-VDIS is not fast enough, we design an efficient exact algorithm, called MTG-VDGE, which exploits the degree metric to sort the vertexes and proposes a new combination order, namely degree and reverse based branch and bound (DRBB). MTG-VDGE gives priority to those vertices with small degree. For a large p, we further develop an approximation algorithm, namely MTG-VDLT, which discards candidate attendees with high degree to reduce the number of vertices to be considered. The experimental results on real datasets manifest that the proposed algorithms outperform existing approaches on both efficiency and group tenuity.

  • Image and Graphics
  • RESEARCH ARTICLE
    Jian ZHANG, Fazhi HE, Yansong DUAN, Shizhen YANG

    The haze phenomenon seriously interferes the image acquisition and reduces image quality. Due to many uncertain factors, dehazing is typically a challenge in image processing. The most existing deep learning-based dehazing approaches apply the atmospheric scattering model (ASM) or a similar physical model, which originally comes from traditional dehazing methods. However, the data set trained in deep learning does not match well this model for three reasons. Firstly, the atmospheric illumination in ASM is obtained from prior experience, which is not accurate for dehazing real-scene. Secondly, it is difficult to get the depth of outdoor scenes for ASM. Thirdly, the haze is a complex natural phenomenon, and it is difficult to find an accurate physical model and related parameters to describe this phenomenon. In this paper, we propose a black box method, in which the haze is considered an image quality problem without using any physical model such as ASM. Analytically, we propose a novel dehazing equation to combine two mechanisms: interference item and detail enhancement item. The interference item estimates the haze information for dehazing the image, and then the detail enhancement item can repair and enhance the details of the dehazed image. Based on the new equation, we design an anti-interference and detail enhancement dehazing network (AIDEDNet), which is dramatically different from existing dehazing networks in that our network is fed into the haze-free images for training. Specifically, we propose a new way to construct a haze patch on the flight of network training. The patch is randomly selected from the input images and the thickness of haze is also randomly set. Numerous experiment results show that AIDEDNet outperforms the state-of-the-art methods on both synthetic haze scenes and real-world haze scenes.

  • RESEARCH ARTICLE
    Bo WANG, Zitong KANG, Pengwei DONG, Fan WANG, Peng MA, Jiajing BAI, Pengwei LIANG, Chongyi LI

    Underwater images often exhibit severe color deviations and degraded visibility, which limits many practical applications in ocean engineering. Although extensive research has been conducted into underwater image enhancement, little of which demonstrates the significant robustness and generalization for diverse real-world underwater scenes. In this paper, we propose an adaptive color correction algorithm based on the maximum likelihood estimation of Gaussian parameters, which effectively removes color casts of a variety of underwater images. A novel algorithm using weighted combination of gradient maps in HSV color space and absolute difference of intensity for accurate background light estimation is proposed, which circumvents the influence of white or bright regions that challenges existing physical model-based methods. To enhance contrast of resultant images, a piece-wise affine transform is applied to the transmission map estimated via background light differential. Finally, with the estimated background light and transmission map, the scene radiance is recovered by addressing an inverse problem of image formation model. Extensive experiments reveal that our results are characterized by natural appearance and genuine color, and our method achieves competitive performance with the state-of-the-art methods in terms of objective evaluation metrics, which further validates the better robustness and higher generalization ability of our enhancement model.

  • Information Security
  • LETTER
    Zhongqiu WANG, Shixiong XIA, Fengrong ZHANG