检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Qiu, Zhen Zhang, Yifan Lin, Hongbin Niu, Shuaicheng Liu, Yanxia Du, Qing Tan, Mingkui School of Software Engineering South China University of Technology School of Computing National University of Singapore Singapore Key Laboratory of Big Data and Intelligent Robot Ministry of Education Pazhou Laboratory

We study a practical domain adaptation task, called source-free unsupervised domain adaptation (UDA) problem, in which we cannot access source domain data due to data privacy issues but only a pre-trained source model and unlabeled target data are available. This task, however, is very difficult due to one key challenge: the lack of source data and target domain labels makes model adaptation very challenging. To address this, we propose to mine the hidden knowledge in the source model and exploit it to generate source avatar prototypes (i.e., representative features for each source class) as well as target pseudo labels for domain alignment. To this end, we propose a Contrastive Prototype Generation and Adaptation (CPGA) method. Specifically, CPGA consists of two stages: (1) prototype generation: by exploring the classification boundary information of the source model, we train a prototype generator to generate avatar prototypes via contrastive learning. (2) prototype adaptation: based on the generated source prototypes and target pseudo labels, we develop a new robust contrastive prototype adaptation strategy to align each pseudo-labeled target data to the corresponding source prototypes. Extensive experiments on three UDA benchmark datasets demonstrate the effectiveness and superiority of the proposed method. Copyright © 2021, The Authors. All rights reserved.

关键词： data privacy

来源：评论

学校读者我要写书评

暂无评论

CLDG: Contrastive Learning on Dynamic Graphs

arXiv

引用

arXiv 2024年

作者： Xu, Yiming Shi, Bin Ma, Teng Dong, Bo Zhou, Haoyi Zheng, Qinghua Department of Computer Science and Technology Xi’an Jiaotong University China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University China Department of Distance Education Xi’an Jiaotong University China School of Software Beihang University China Advanced Innovation Center for Big Data and Brain Computing Beihang University China

The graph with complex annotations is the most potent data type, whose constantly evolving motivates further exploration of the unsupervised dynamic graph representation. One of the representative paradigms is graph contrastive learning. It constructs self-supervised signals by maximizing the mutual information between the statistic graph’s augmentation views. However, the semantics and labels may change within the augmentation process, causing a significant performance drop in downstream tasks. This drawback becomes greatly magnified on dynamic graphs. To address this problem, we designed a simple yet effective framework named CLDG. Firstly, we elaborate that dynamic graphs have temporal translation invariance at different levels. Then, we proposed a sampling layer to extract the temporally-persistent signals. It will encourage the node to maintain consistent local and global representations, i.e., temporal translation invariance under the timespan views. The extensive experiments demonstrate the effectiveness and efficiency of the method on seven datasets by outperforming eight unsupervised state-of-the-art baselines and showing competitiveness against four semi-supervised methods. Compared with the existing dynamic graph method, the number of model parameters and training time is reduced by an average of 2,001.86 times and 130.31 times on seven datasets, respectively. The code and data are available at: https://***/yimingxu24/CLDG. © 2024, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Calibration Signal Assisted Emitter Localization under Sensor Position Uncertainty

引用

IEEE Transactions on Aerospace and Electronic Systems 2025年

作者： You, Ming-Yi Pu, Wenqiang Zhou, Rui Ye, Yun-Xia Xiao, Minrui Wang, Wei Shi, Qingjiang No.36 Research Institute of CETC National Key Laboratory of Electromagnetic Space Security Jiaxing314033 China The Chinese University of Hong Kong-Shenzhen Shenzhen Research Institute of Big Data Shenzhen518172 China Shanghai Jiaotong University School of Sensing Science and Engineering Shanghai200240 China Tongji University School of Software Engineering Shanghai200070 China

The uncertainty of sensor position significantly compromises the accuracy of source localization within the direct position determination (DPD) framework. This paper explores the use of an external calibrator that emits a calibration signal with a known position to mitigate the impact of sensor position uncertainty on localization accuracy. By utilizing the calibration signal, we develop a refined Gibbs sampling algorithm as the initial step to effectively address the high-dimensional optimization problem associated with sensor position calibration. In the second step, we account for measurement noise and estimation uncertainties in the calibrated sensor positions and propose a revised DPD method, which demonstrates greater resilience to sensor position uncertainty compared to existing methods. We also derive the Cramer-Rao Lower Bound for emitter position under scenarios both with and without an external calibrator. Numerical simulations are performed to evaluate the performance of our proposed algorithms against several benchmark methods, demonstrating the effectiveness of our approach. © 1965-2011 IEEE.

关键词： Cramer-Rao bounds

来源：评论

学校读者我要写书评

暂无评论

DarkSAM: Fooling Segment Anything Model to Segment Nothing

arXiv

引用

arXiv 2024年

作者： Zhou, Ziqi Song, Yufei Li, Minghui Hu, Shengshan Wang, Xianlong Zhang, Leo Yu Yao, Dezhong Jin, Hai National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Cluster and Grid Computing Lab China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

Segment Anything Model (SAM) has recently gained much attention for its outstanding generalization to unseen data and tasks. Despite its promising prospect, the vulnerabilities of SAM, especially to universal adversarial perturbation (UAP) have not been thoroughly investigated yet. In this paper, we propose DarkSAM, the first prompt-free universal attack framework against SAM, including a semantic decoupling-based spatial attack and a texture distortion-based frequency attack. We first divide the output of SAM into foreground and background. Then, we design a shadow target strategy to obtain the semantic blueprint of the image as the attack target. DarkSAM is dedicated to fooling SAM by extracting and destroying crucial object features from images in both spatial and frequency domains. In the spatial domain, we disrupt the semantics of both the foreground and background in the image to confuse SAM. In the frequency domain, we further enhance the attack effectiveness by distorting the high-frequency components (i.e., texture information) of the image. Consequently, with a single UAP, DarkSAM renders SAM incapable of segmenting objects across diverse images with varying prompts. Experimental results on four datasets for SAM and its two variant models demonstrate the powerful attack capability and transferability of DarkSAM. Our codes are available at: https://***/CGCL-codes/DarkSAM. Copyright © 2024, The Authors. All rights reserved.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

Advances in Robust Federated Learning: A Survey with Heterogeneity Considerations

arXiv

引用

arXiv 2024年

作者： Chen, Chuan Liao, Tianchi Deng, Xiaojun Wu, Zihou Huang, Sheng Zheng, Zibin School of Computer Science and Engineering Sun Yat-sen University Guangzhou510006 China Guangdong Key Laboratory of Big Data Analysis and Processing China School of Software Engineering Sun Yat-sen University Zhuhai519000 China Zhuhai Key Laboratory of Trusted Large Language Models China School of Systems Science and Engineering Sun Yat-sen University Guangzhou510006 China

In the field of heterogeneous federated learning (FL), the key challenge is to efficiently and collaboratively train models across multiple clients with different data distributions, model structures, task objectives, computational capabilities, and communication resources. This diversity leads to significant heterogeneity, which increases the complexity of model training. In this paper, we first outline the basic concepts of heterogeneous FL and summarize the research challenges in FL in terms of five aspects: data, model, task, device and communication. In addition, we explore how existing state-of-the-art approaches cope with the heterogeneity of FL, and categorize and review these approaches at three different levels: data-level, model-level, and architecture-level. Subsequently, the paper extensively discusses privacy-preserving strategies in heterogeneous FL environments. Finally, the paper discusses current open issues and directions for future research, aiming to promote the further development of heterogeneous FL. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

DRL-Based Time-Varying Workload Scheduling With Priority and Resource Awareness

引用

IEEE Transactions on Network and Service Management 2025年第3期22卷 2838-2852页

作者： Liu, Qifeng Fan, Qilin Zhang, Xu Li, Xiuhua Wang, Kai Xiong, Qingyu Chongqing University School of Big Data and Software Engineering Chongqing400044 China Chongqing University Key Laboratory of Dependable Service Computing in Cyber Physical Society of Ministry of Education Chongqing400044 China Nanjing University School of Electronic Science and Engineering Nanjing210023 China Haihe Laboratory of Information Technology Application Innovation Tianjin300072 China Harbin Institute of Technology School of Computer Science and Technology Weihai264209 China Shandong Key Laboratory of Industrial Network Security Weihai264209 China

With the proliferation of cloud services and the continuous growth in enterprises' demand for dynamic multi-dimensional resources, the implementation of effective strategy for time-varying workload scheduling has become increasingly significant. In this paper, we propose a deep reinforcement learning (DRL)-based method for time-varying workload scheduling, aiming to allocate resources efficiently across servers in the cluster. Specifically, we integrate a classifier and queue scorer to construct a priority queue that exploits temporal resource utilization patterns across different workload classes. Then, we design parallel graph attention layers to capture the dimensional features and temporal dynamics of cloud server cluster. Moreover, we propose a DRL algorithm to generate scheduling strategies that can adapt to dynamic environments. Validation on real-world traces from Google cluster demonstrates that our method outperforms existing approaches in key metrics of cloud server cluster management. © 2004-2012 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Incorporating entity type information into knowledge representation learning 5

Incorporating entity type information into knowledge represe...

引用

5th IEEE International Conference on data Science in Cyberspace, DSC 2020

作者： Huang, Wenyu Wang, Guohua Zhang, Huakui Chen, Feng South China University of Technology School of Software Engineering Guangzhou China South China University of Technology Ministry of Education Key Laboratory of Big Data and Intelligent Robot Guangzhou China

ISBN: (纸本)9781728195582

Knowledge Representation Learning (KRL), which is also known as Knowledge Embedding, is a very useful method to represent complex relations in knowledge graphs. The low-dimensional representation learned by KRL models makes a contribution to many tasks like recommender system and question answering. Recently, many KRL models are trained using square loss or cross entropy loss based on Closed World Assumption (CWA). Although CWA is an easy way for training, it violates the link prediction task which exploits KRL. To overcome the drawback, in this paper, we introduce a new method, Type-based Prior Possibility Assumption (TPPA). TPPA calculates type based prior possibilities for missing triplets instead of zeros in the training process of KRL to weaken the bad influence of CWA. We compare TPPA with the baseline method CWA in ConvE and TuckER, two common frameworks for knowledge representation learning. The experiment results on FB15k-237 dataset show that TPPA based training method outperforms CWA based training method in link prediction task. © 2020 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Towards Distractibility Induced trust Management Using BlockChain for Edge Computing

Towards Distractibility Induced trust Management Using Block...

引用

International Conference on Image, Vision and intelligent Systems, ICIVIS 2021

作者： Yang, Haochen Wang, Guanghui Dong, Lifeng He, Xin School of Software Henan University Kaifeng China Henan Provincial Engineering Research Center of Intelligent Data Processing Kaifeng China Henan International Joint Laboratory of Intelligent Network Theory and Key Technology Henan University Kaifeng China Institute of Intelligent Network System Henan University Kaifeng China Henan Jiuyu Tenglong Information Engineering Co. Ltd. Zhengzhou China

ISBN: (纸本)9789811669620

Edge computing solves massive data processing problems by utilizing the storage and computing functions of edge devices. Due to the openness of edge computing, distracted devices inevitably participate in edge computing and affect the network computing performance, which degrades trust management of edge devices. To solve the above issue, this paper proposes a trust management framework based on blockchain for edge computing scenarios (ECTMF). First, the ECTMF framework slices users and provides corresponding services based on different requirements. It evaluates the trustworthiness of end devices and distinguishes distracted devices in edge computing. Then, the Proof of Monitor (PoM) consensus mechanism is proposed to continuously monitor distracted end devices in edge computing. Simulations validate the effectiveness and efficiency of the ECTMF and PoM. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

WMAJL: Watcher-Mediated Attention Joint Learning Model for Multimodal Relation Extraction

WMAJL: Watcher-Mediated Attention Joint Learning Model for M...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yunrui Dong Guiduo Duan Tianxi Huang Yunhao Li School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Laboratory of Intelligent Collaborative Computing University of Electronic Science and Technology of China Chengdu China Trusted Cloud Computing and Big Data Key Laboratory of Sichuan Province Chengdu China Chengdu Textile College College of Humanities and General Education Chengdu China School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In the domain of Multimodal Relation Extraction (MRE), we present the $\color{Red}{\text{W}}$atcher-$\color{Red}{\text{M}}$ediated $\color{Red}{\text{A}}$ttention $\color{Red}{\text{J}}$oint $\color{Red}{\text{L}}$earning Model ($\color{Red}{\text{WMAJL}}$), a novel approach addressing the challenges of modality alignment noise, cross-modal fusion disparity, preservation of textual relative position information, and the distinctiveness of classification labels. WMAJL employs an integrative framework leveraging contrastive learning and variational autoencoder constraints to mitigate modality alignment noise by prioritizing relevant semantic data and effectively reducing extraneous noise that does not contribute to the task. The model’s innovative architecture includes a mediator watcher, which facilitates enhanced cross-modal fusion by enabling nuanced information exchange between textual and visual modalities while preserving the unique characteristics of each modality. Additionally, the design of auxiliary tasks, such as Named Entity Recognition (NER), and output supervision constructs loss functions that preserve relative position information, ensuring a precise depiction of entity relationships throughout the multilayer encoding processes. A key differentiator of WMAJL is its label-centric self-information loss technique, inspired by InfoNCE, which trains the model to cluster similar relation labels in semantically coherent areas, thereby optimizing classification label uniqueness by discerning subtle differences among relation types. The synergistic application of these strategies has led to a significant enhancement of WMAJL’s performance, as evidenced by its state-of-the-art F1 score of $\color{Red}{84.93\%}$ on the MNRE dataset. This achievement surpasses existing benchmarks and sets a new standard for multimodal knowledge extraction, underscoring WMAJL’s potential to revolutionize the MRE landscape.

关键词： Resistance Visualization Technological innovation Noise Autoencoders Semantics Contrastive learning Nonhomogeneous media Speech processing Standards

来源：评论

学校读者我要写书评

暂无评论

Look from Coarse to Fine: A Hierarchical Lane Detection Network

Look from Coarse to Fine: A Hierarchical Lane Detection Netw...

引用

Artificial Intelligence Innovation (ICAII), International Conference on

作者： Xichen Yu Yanduo Zhang Tao Lu Xun Li Jun Chang Hubei Provincial Key Laboratory of Intelligent Robot School of Computer Science and Engineering & School of Artificial Intelligence Wuhan Institute of Technology Wuhan China Hubei Key Laboratory of Multimedia and Network Communication Engineering National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence and School of Computer Science Wuhan University Wuhan China

Lane detection involves identifying a high-level semantic identifier with a slender structure, which requires both high-level feature detection for existence and low-level features for shape and position determination. The objective of this study is to introduce a framework called hierarchical lane detection (HLDNet), which from coarse to fine established stronger relationships between the global and local lanes to achieve stable and accurate lane detection. Specifically, we reframe the existing lane identification method by expressing lanes as a combination of lane instances (for existence detection) and lane point sets (for shape detection). Our proposed Lane IoU loss improves the model performance by regressing the lane instance as a whole. To enhance stability, we capture the complete lane with high-level features using multiple receptive fields. Additionally, our model adaptively fuses high- and low-level features, propagates key features between channels, and further enhances lane local feature representation. Our experimental results show significant improvements in the stability and accuracy of lane detection across multiple benchmark datasets and can run in realtime at a speed of 91 FPS.

关键词：

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

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

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

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

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：