检索结果-内蒙古大学图书馆

Regularized Total Least Squares Broad Learning System for Regression

Journal of Computers (Taiwan) 2025年第1期36卷 89-112页

作者： Xiong, Ke-Jia Yang, Guan-Ci Zhou, Tao Xie, Zhen-Qiang State Key Laboratory of Public Big Data Guizhou University Guizhou Province Guiyang City550025 China Key Laboratory of Advanced Manufacturing Technology of the Ministry of Education Guizhou University Guizhou Province Guiyang City550025 China National Engineering Research Center of Big Data Application to The Improvement of Governance Capacity Guizhou Province Guiyang City550000 China

The Broad Learning System (BLS) has been extensively developed and applied across various fields due to its significant advantages, including high efficiency, strong generality, and scalability. However, in practical applications where both inputs and outputs are contaminated by noise, traditional BLS demonstrates suboptimal performance in handling sample data. This study introduces a novel regularized total least squares broad learning system (RTLS-BLS) designed to enhance the robustness of BLS when noisy values are present in both the input and the output of the training set. Initially, the k-support norm is integrated into the BLS-based autoencoder (BLS-AE), embedded within BLS, to extract robust features from the original input data. The BLS-AE equipped with the k-support norm effectively addresses or mitigates issues related to overly sparse (based on L1-norm) or overly dense (based on L2-norm) regularization of extracted features. Subsequently, regularized total least squares (RTLS) are employed to assess the output weights of BLS, further enhancing its robustness. Moreover, simultaneous perturbation measures for the coefficient matrix and the output are provided. Experimental results from two function approximation tasks, eight benchmark regression tasks, and two network interface flow datasets demonstrate that the proposed RTLS-BLS achieves significantly robust performance under noisy conditions compared to other methods. © 2025 Codon Publications. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Cooperative Eco-Driving for Mixed Platoons with Heterogeneous Energy at a Signalized Intersection Based on a Mixed Space-Time-State Network

引用

IEEE Transactions on Vehicular technology 2025年

作者： Xiao, Guosheng Wu, Yunxia Jiang, Yangsheng Ran, Bin Yao, Zhihong Southwest Jiaotong University School of Transportation and Logistics National Engineering Laboratory of Integrated Transportation Big Data Application Technology National United Engineering Laboratory of Integrated and Intelligent Transportation Sichuan610031 China University of Wisconsin-Madison Department of Civil and Environmental Engineering MadisonWI53706 United States

With the ongoing trend of vehicle automation and electrification, the future will present a mixed traffic environment with different automation and energy consumption types. However, current research focuses more on fully connected or homogeneous energy consumption scenarios in eco-driving, ignoring the impact of human-driven vehicles and electric vehicles (EVs) with high energy efficiency and regenerative braking capabilities. This study proposes a cooperative eco-driving framework for mixed vehicle platoons with heterogeneous energy at a signalized intersection. A Mixed Space-Time-State Network is developed to optimize the trajectory of connected autonomous vehicles (CAVs), while predicting and describing the trajectories of human-driven vehicles. Then, a modified dynamic programming algorithm is formulated, which converts the highly nonlinear optimization problem into a state-space search problem. The solving efficiency is further improved by 85% with a feasible state space reduction algorithm, which can meet the real-time computational requirements. The result indicates that cooperation within the mixed platoon can smooth out the trajectory while considering the composition of energy consumption. In addition, the energy efficiency can be improved by 20.39% with a low CAV penetration rate. © 1967-2012 IEEE.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Robust Multi-Object Tracking Using Vision Sensor with Fine-Grained Cues in Occluded and Dynamic Scenes

引用

IEEE Sensors Journal 2025年第11期25卷 20547-20560页

作者： Hu, Yaoqi Sun, Jinqiu Jin, Hao Niu, Axi Yan, Qingsen Zhu, Yu Zhang, Yanning School of Computer Science China School of Astronautics China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Northwestern Polytechnical University Dongxiang Road Shaanxi Xi'an710129 China

Multi-object tracking (MOT) using vision sensors remains a challenging problem, particularly in dynamic backgrounds and severe occlusions. Existing methods, relying on holistic appearance or spatial cues, fail to capture detailed information within object regions and the background, resulting in inaccurate and inconsistent tracking. To address these issues, we propose a novel Detail-Driven Multi-Object Tracking (DD-MOT) method that revisits and leverages fine-grained cues to discover the dynamics of both object regions and the background, facilitating the recovery and association of trajectories. Specifically, the proposed method consists of three key modules: (i) Points Trajectories Generator, (ii) Camera Motion and Occlusion Compensation, and (iii) Fine- and Coarsegrained Association. The Points Trajectories Generator is responsible for generating fine-grained cues by sampling an initial set of points, generating point trajectories, and refining the initial set of points. The Camera Motion and Occlusion Compensation module utilizes the background and object point trajectories to correct background motion and recover occluded object bounding boxes. Finally, the Fine- and Coarse-grained Association module leverages point trajectory cues to assist in establishing a more effective association strategy, combining both fine-grained and coarse-grained spatial cues from the object bounding boxes. We evaluate our DD-MOT method on several benchmark datasets, including MOT17, MOT20, demonstrating that it consistently outperforms state-of-the-art (SOTA) methods in key metrics such as HOTA (64.2, 62.6), MOTA (80.8, 78.1), and IDF1 (78.8, 77.1). © 2025 IEEE.

关键词： Motion compensation

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Based Proactive Anomaly Detection for 5G Core Control Plane Network Function Interactions

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Tan, Yawen Liu, Jiajia Li, Yuanhao Wang, Jiadai Xidian University State Key Laboratory of Integrated Services Networks School of Cyber Engineering Xi'an710071 China Northwestern Polytechnical University National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Cybersecurity Shaanxi Xi'an710072 China

The Service-Based Architecture (SBA) introduced by 3GPP allows the control plane of 5G Core Network (CN) to function through a set of interconnected Network Functions (NFs), which offers significant benefits such as improved scalability, simplified operations, and efficient resource utilization. However, such a new transformation could also make critical NFs in 5G CN become more susceptible to attacks due to both internal vulnerabilities within the software-implemented NFs or the expanded interfaces brought by the SBA. Existing studies targeting 5G network security pay little attention to the detection of abnormal behaviors from control plane NFs. Therefore, in this paper, we propose 5GCGuard, an anomaly detection scheme aiming at identifying abnormal NF interactions in 5G CN. It employs a deep learning-based sequence model to learn normal interaction patterns and detects anomalies based on deviations from the learned pattern. Special designs including attention mechanism, multi-task learning, probabilistic labeling and automatic threshold decision are made to enhance its detection capabilities. We establish a cloud-native 5G testbed to evaluate the effectiveness of 5GCGuard. Extensive evaluation results show its superiority over comparison schemes. Case studies also demonstrate its ability to proactively detect various NF interaction anomalies, enabling swift action to block attacks and prevent further damage. © 2015 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

On time slices, time complexity, and energy increase in implementing quantum adiabatic evolution by circuit model

引用

European Physical Journal D 2025年第5期79卷 52-52页

作者： Sun, Jie Qian, Ling Cai, Dunbo Huang, Zhiguo Lu, Songfeng School of Internet Anhui University Hefei230039 China National Engineering Research Center of Agro-Ecological Big Data Analysis and Application Anhui University Hefei230601 China Fin China-Anhui University Joint Laboratory for Financial Big Data Research Hefei Financial China Information & Technology Co. Ltd. Hefei230022 China Software Technology Company Limited Suzhou215163 China School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan430074 China

Abstract: In this paper, motivated by previous related works on a kind of quantum local adiabatic evolution, we mainly study the circuit model of its corresponding quantum global adiabatic search algorithm, in which the evolving paths functions in the system Hamiltonian have a general form instead of the usual linear interpolating ones which are widely investigated in the literature. The main conclusion found here can be summarized as follows. Though this quantum adiabatic computation can demonstrate various algorithmic performances, when considering implementing it by a quantum circuit, the time slices thus produced always keep as an invariant which is proportional to inverse square of the overlap between the initial and final states in the problem of interest and also equal to that of multiplying the time complexity and the energy together, by noting that compared with the usual adiabatic evolutions, there is being a great increase of energy of the system. As far as we know, the result exposed here is new and is hoped to be useful for further understanding the kind of quantum adiabatic evolution itself and its connection with the quantum circuit model. Also, we hope the main conclusion here will be ubiquitous when considering applying this kind of quantum adiabatic evolution to solve the problems beyond quantum search. Graphic abstract: (Figure presented.) © The Author(s), under exclusive licence to EDP Sciences, SIF and Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Hamiltonians

来源：评论

学校读者我要写书评

暂无评论

The Death and Life of Free-Floating Car Sharing in China: Case Study of Chengdu and Changchun 1st

The Death and Life of Free-Floating Car Sharing in China: Ca...

引用

1st International Symposium on Intelligent technology for Future Transportation, ITFT 2024

作者： Wang, Hongjie Luo, Xia Su, Qiming Bao, Hongqing School of Transportation and Logistics Southwest Jiaotong University Chengdu China National Engineering Laboratory of Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu China National United Engineering Laboratory of Integrated and Intelligent Transportation Southwest Jiaotong University Chengdu China

ISBN: (纸本)9783031841477

In the last decades, free-floating car sharing (FFCS) has become a fast-growing mode of mobility service and promoted in many cities across the world. While lots of researchers have paid increasing attention to FFCS in recent years, little is known regarding the characterization and evolution of FFCS in Chinese cities, where car sharing services are gaining more and more popularity. In this work, we focus on two Chinese cities: Chengdu and Changchun, both of which have witnessed the shutdown of FFCS service and its replacement with station-based car sharing (SBCS). We rebuild 743,796 trips using public data from the smartphone application of a car sharing service, then explore the characteristics and spatiotemporal distribution of FFCS usage, and identify the differences compared to those in Europe and North America. Furthermore, we conduct a comparative analysis of the service coverage, demand spatiotemporal distribution and usage patterns of car sharing services before and after the shutdown of FFCS. Results show that, the shutdown of FFCS did not cause a severe disruption to the car sharing service. Instead, the introduction of SBCS effectively inherited and accommodated the travel demand. This work underscores the importance of considering the unique characteristics of China’s FFCS services when develop operation strategies and demonstrates that a well-designed SBCS could emerge as a viable alternative. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： big data Operation Scheme Transition Shared Mobility

来源：评论

学校读者我要写书评

暂无评论

Salix-Leaf: Find Main Veins of Signal Clusters for Practical Parallel Decoding

引用

IEEE Transactions on Mobile Computing 2025年

作者： Li, Yajun Zhao, Jumin Li, Dengao Wu, Hejun Xu, Shuang Bai, Ruiqin Taiyuan University of Technology College of Electronic Information and Optical Engineering Taiyuan China Key Laboratory of Big Data Fusion Analysis and Application of Shanxi Province Taiyuan China Intelligent Perception Engineering Technology Center of Shanxi Taiyuan China Taiyuan University of Technology College of Computer Science and Technology College of Data Science Taiyuan China Sun Yat-sen University Department of Computer Science Guangzhou China

Parallel decoding of backscatter improves communication throughput by enabling concurrent transmission of backscatter tags. In practical applications of parallel decoding, it is extremely difficult to distinguish collided signals in superclusters where multiple signal clusters overlap. Existing methods are usually effective for superclusters with uniformly distributed signals. Nevertheless, there are many more scenarios in which signals in superclusters tend to gather unevenly, and existing methods cannot work. Such uneven clustering of signals occurs due to the following two possible causes: (1) signal-strengthdifferences (SSDs) among tags;or (2) cluster drifting (CD) driven by interferences from other objects within communication environments. This paper proposes a novel scheme called SalixLeaf, which aims to identify the main veins of signal clusters to address this problem of superclusters with unevenly distributed signals. Salix-Leaf identifies the main vein of each signal cluster for fine-grained clustering so that the direction of the main veins can be used to verify the accuracy of clustering. In addition, SalixLeaf employs a supercluster decomposer that divides signals into different segments for clustering analysis, enhancing robustness and practicability. Experimental results show that Salix-Leaf achieves a 1.2-fold increase in throughput and a 25% reduction in bit error rate (BER) compared to the state-of-the-art. © 2002-2012 IEEE.

关键词： Backscattering

来源：评论

学校读者我要写书评

暂无评论

Segment Together: A Versatile Paradigm for Semi-Supervised Medical Image Segmentation

引用

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Zeng, Qingjie Xie, Yutong Lu, Zilin Lu, Mengkang Wu, Yicheng Xia, Yong Northwestern Polytechnical University National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Xi’an710072 China The University of Adelaide Australian Institute for Machine Learning AdelaideSA5000 Australia Monash University Faculty of Information Technology Department of Data Science and AI Australia

The scarcity of annotations has become a significant obstacle in training powerful deep-learning models for medical image segmentation, limiting their clinical application. To overcome this, semi-supervised learning that leverages abundant unlabeled data is highly desirable to enhance model training. However, most existing works still focus on specific medical tasks and underestimate the potential of learning across diverse tasks and datasets. In this paper, we propose a Versatile Semi-supervised framework (VerSemi) to present a new perspective that integrates various SSL tasks into a unified model with an extensive label space, exploiting more unlabeled data for semi-supervised medical image segmentation. Specifically, we introduce a dynamic task-prompted design to segment various targets from different datasets. Next, this unified model is used to identify the foreground regions from all labeled data, capturing cross-dataset semantics. Particularly, we create a synthetic task with a CutMix strategy to augment foreground targets within the expanded label space. To effectively utilize unlabeled data, we introduce a consistency constraint that aligns aggregated predictions from various tasks with those from the synthetic task, further guiding the model to accurately segment foreground regions during training. We evaluated our VerSemi framework against seven established SSL methods on four public benchmarking datasets. Our results suggest that VerSemi consistently outperforms all competing methods, beating the second-best method with a 2.69% average Dice gain on four datasets and setting a new state of the art for semi-supervised medical image segmentation. Code is available at https://***/maxwell0027/VerSemi. © 1982-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Out-of-Distribution Fault Diagnosis of Industrial Cyber-Physical Systems Based on Orthogonal Anchor Clustering With Adaptive Balance

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical Systems 2025年 3卷 48-60页

作者： Liu, Ruonan Hu, Puyuan Zhao, Siheng Sun, Zhijian Han, Te Pang, Zhibo Zhang, Weidong Shanghai Jiao Tong University Department of Automation Shanghai200240 China National Key Laboratory of Marine Engine Science and Technology Shanghai201108 China Crossocean of Suzhou Technology Suzhou215121 China Beijing Institute of Technology Center for Energy and Environmental Policy Research School of Management Beijing100081 China ABB Corporate Research Sweden Department of Automation Technology Vasteras722 26 Sweden KTH Royal Institute of Technology Department of Intelligent Systems Stockholm114 28 Sweden Henan University Henan Key Laboratory of Big Data Analysis and Processing School of Computer and Information Engineering Kaifeng475001 China Shanghai Jiaotong University Department of Automation Shanghai200240 China

Given the critical role of rotating machinery in industrial cyber-physical systems (ICPS), ensuring their reliable operation is essential for the stability and safety of ICPS. Deep neural networks have demonstrated competitive results for intelligent fault diagnosis, which are usually trained via the historical data of all fault modes. However, in real engineering, it is usually difficult to collect samples and exhaust all failures during the training stage. As a result, out-of-distribution fault diagnosis (OOD-FD) becomes a more realistic problem that requires the methods to not only accurately diagnose the known faults, but also effectively recognize unknown ones. Therefore, a novel orthogonal anchor clustering with focal attention (FA-OAC) is proposed in this paper for OOD-FD. Firstly, an orthogonal anchor clustering (OAC) algorithm is proposed to fix the class center of each fault mode orthogonally and distinguish the known and unknown faults at the class level. Then, because the identifiability of different fault modes changes a lot in OOD-FD, the focal attention mechanism is applied to dynamically adjust the attention to different fault modes according to the distance loss of OAC, thus addressing the identifiability imbalance problem. To verify the effectiveness of the proposed method, different OOD-FD tasks are designed based on two rotating machinery datasets. The experimental results and comparison with state-of-the-art fault diagnosis methods have demonstrated that the proposed method has improved the OOD-FD performance greatly and therefore provides an effective tool in real engineering. © 2023 IEEE.

关键词： Rotating machinery

来源：评论

学校读者我要写书评

暂无评论

Predicting CRISPR Cas9 Off-Target Activities With a Two-Stage Deep Learning Framework

IEEE Transactions on Computational Biology and Bioinformatic...

引用

IEEE Transactions on Computational Biology and Bioinformatics 2025年第2期22卷 956-962页

作者： Tianshan Zhang Bei Jiang Guang Yang Zhijie Bao Jiajie Peng AI for Science Interdisciplinary Research Center School of Computer Science Northwestern Polytechnical University Xi'an China Key Laboratory of Big Data Storage and Management Ministry of Industry and Information Technology Northwestern Polytechnical University Shanxi China The National Engineering Laboratory for In tegrated Aerospace Ground Ocean Big Data Application Technology Shanxi China Tianjin second people's hospital Tianjin China Research and Development Institute of Northwestern Polytechnical University in Shenzhen Shenzhen China

The CRISPR-Cas9 system, found across bacteria and archaea, enables efficient genome engineering in eukaryotic cells. There is a challenge that Cas9 guide RNA may cause off-target activities. Although numbers of methods have been proposed to predict off-target activities in guide RNA designing, this procedure involves numerous potential off-target sites, which causes label imbalance problem. To address this problem, we developed a deep learning framework, named CAF-Net (Cas9 Augmentation and Finetune Network), for predicting off-target activities of CRISPR-Cas9. First, we pretrain an embedding model to extract features from target and guide sequence pairs. Subsequently, data augmentation is applied to these features. And finally, the model is finetuned with synthetic samples. Evaluation results demonstrate its performance on published datasets.

关键词： RNA Bioinformatics Feature extraction DNA Genomics Convolutional neural networks Computational modeling Transformers Predictive models CRISPR

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：