检索结果-内蒙古大学图书馆

作者： Chen, Haibo Chen, Jinfu Wu, Yucheng Cai, Saihua Ahmad, Bilal Huang, Rubing Wang, Shengran Zhang, Chi School of Computer Science and Communication Engineering Jiangsu Key Laboratory of Security Technology Industrial Cyberspace Jiangsu University Zhenjiang China School of Computer Science and Engineering Faculty of Innovation Engineering Macau University of Science and Technology China

Deep neural networks (DNNs) possess potent feature learning capability, enabling them to comprehend natural language, which strongly support developing dialogue systems. However, dialogue systems usually perform incorrect behaviours in some corner cases, which may cause misunderstanding or economic loss. To test and debug dialogue systems, a popular fuzzing framework by metamorphic testing with Gini impurity guidance is proposed, namely, DialTest. However, DialTest treats all seeds (the initial test inputs to generate the mutated test inputs) equally during the fuzzing process and does not differentiate seeds, resulting in a certain limitation to its incorrect behaviour detection capability. In this paper, we propose to enhance the DialTest by applying a lightweight energy adjustment strategy called DialTest with Energy Adjustment (DialTest-EA). DialTest-EA employs the ant colony optimization algorithm (ACO) to adjust the mutation energy of each seed adaptively, ensuring that potential seeds have more opportunities to generate subsequent test inputs. To evaluate the effectiveness of the proposed DialTest-EA, we conduct a series of comparisons with the original DialTest and random mutation strategy. The experimental results show that the proposed DialTest-EA outperforms the compared methods both in the intent detection and slot filling tasks. Compared with the original DialTest, the intent detection accuracy of generated test cases by the proposed method is reduced by more than 14%, and the slot filling accuracy is reduced by more than 8%. © 2024 John Wiley & Sons Ltd.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Lightweight Self-Supervised Monocular Depth Estimation for All-Day Scenes Using Generative Adversarial Network

Lightweight Self-Supervised Monocular Depth Estimation for A...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Junding Rao, Di Akoudad, Youssef Gao, Wei Chen, Jie School of Computer Science and Communication Engineering Jiangsu University China CIAIC School of Marine Science and Technology Northwestern Polytechnical University China

ISBN: (纸本)9798350368741

Self-supervised monocular depth estimation (MDE) has achieved performance levels comparable to supervised methods in well-lit environments. However, current methods struggle particularly with challenging nighttime scenes. Existing all-day self-supervised MDE methods often rely on specialized nighttime datasets, which require extensive data collection and annotation, adding complexity and resource demands to the training process. To overcome this limitation, we propose ADDepth, a novel lightweight all-day self-supervised MDE network. ADDepth leverages CoMoGAN to transform daytime images into nighttime scenes, thereby circumventing the need for a separate nighttime dataset. Additionally, we introduce a low-scale consistency loss to enhance depth map quality by mitigating the issue of blurred depth predictions, a common challenge caused by the reduced number of convolutional kernels in decoder layers. Our approach retains the network's lightweight design while significantly improving its generalization across different lighting conditions. Experimental results on public benchmarks validate the superiority of the proposed ADDepth. The source code is available at https://***/zjdzhou/ADDepth. © 2025 IEEE.

关键词： all-day scenes lightweight monocular depth estimation Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Distributed Distortion-Aware Beamforming Designs for Cell-Free mMIMO Systems

引用

IEEE Journal on Selected Topics in Signal Processing 2025年第2期19卷 381-397页

作者： Liu, Mengzhen Li, Ming Liu, Rang Liu, Qian Dalian University of Technology School of Information and Communication Engineering Dalian116024 China University of California Center for Pervasive Communications and Computing IrvineCA92697 United States Dalian University of Technology School of Computer Science and Technology Dalian116024 China

Cell-free massive multi-input multi-output (CF-mMIMO) systems have emerged as a promising paradigm for next-generation wireless communications, offering enhanced spectral efficiency and coverage through distributed antenna arrays. However, the non-linearity of power amplifiers (PAs) in these arrays introduce spatial distortion, which may significantly degrade system performance. This paper presents the first investigation of distortion-aware beamforming in a distributed framework tailored for CF-mMIMO systems, enabling pre-compensation for beam dispersion caused by nonlinear PA distortion. Using a third-order memoryless polynomial distortion model, the impact of the nonlinear PA on the performance of CF-mMIMO systems is firstly analyzed by evaluating the signal-to-interference-noise-and-distortion ratio (SINDR) at user equipment (UE). Then, we develop two distributed distortion-aware beamforming designs based on ring topology and star topology, respectively. In particular, the ring-topology-based fully-distributed approach reduces interconnection costs and computational complexity, while the star-topology-based partially-distributed scheme leverages the superior computation capability of the central processor to achieve improved sum-rate performance. Extensive simulations demonstrate the effectiveness of the proposed distortion-aware beamforming designs in mitigating the effect of nonlinear PA distortion, while also reducing computational complexity and backhaul information exchange in CF-mMIMO systems. © 2025 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

GSA-DT: A Malicious Traffic Detection Model Based on Graph Self-Attention Network and Decision Tree

引用

IEEE Transactions on Network and Service Management 2025年第2期22卷 2059-2073页

作者： Cai, Saihua Tang, Han Chen, Jinfu Lv, Tianxiang Zhao, Wenjun Huang, Chunlei Jiangsu University School of Computer Science and Communication Engineering Zhenjiang212013 China Jiangsu Provincial Key Laboratory of Industrial Cyberspace Security Technology Zhenjiang212013 China

Malicious attack has shown a rapid growth in recent years, it is very important to accurately detect malicious traffic to defend against malicious attacks. Compared with machine learning and deep learning technologies, graph convolutional neural network (GCN) achieves better detection results of malicious traffic due to additional consideration of the correlation between network traffic features. However, existing GCN-based detection models suffer from fixed weight assignment, only focusing on local features, lack the ability to model graph structure and relationships as well as having gradient disappearance. To solve these problems, this paper proposes the GSA-DT model based on graph self-attention network and decision tree. GSA-DT first preprocesses the original network traffic to obtain better traffic features and labels, and then uses GCN to extract the topological structure of network traffic as well as capture the correlation relationships among traffic features, where the ReLU activation function is replaced by LeakyReLU to overcome the problems of neuron "death"and gradient disappearance during the training process;It also introduces the self-attention mechanism into GCN to assign larger weights to the key features to reduce the interference of redundant features. Finally, GSA-DT uses decision tree to perform the detection of malicious traffic. Experimental results on five network traffic datasets show that GSA-DT model improves the detection accuracy over 1% on average than seven advanced malicious traffic detection models, and it also performs better in F1-measure, TPR, FPR as well as stability. © 2004-2012 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep-EERA:DRL-Based Energy-Efficient Resource Allocation in UAV-Empowered Beyond 5G Networks

引用

Tsinghua science and technology 2025年第1期30卷 418-432页

作者： Shabeer Ahmad Jinling Zhang Ali Nauman Adil Khan Khizar Abbas Babar Hayat School of Electronic Engineering Beijing University of Posts and TelecommunicationBeijing 1000876China Department of Information and Communication Engineering Yeungnam UniversityGyeongsan 38541Republic of Korea School of Information Engineering Xi’an Eurasia UniversityXi’an 710065China Department of Computer Science Hanyang UniversitySeoul 04763Republic of Korea

The rise of innovative applications,like online gaming,smart healthcare,and Internet of Things(IoT)services,has increased demand for high data rates and seamless connectivity,posing challenges for Beyond 5G(B5G)*** is a need for cost-effective solutions to enhance spectral efficiency in densely populated areas,ensuring higher data rates and uninterrupted connectivity while minimizing *** Aerial Vehicles(UAVs)as Aerial Base Stations(ABSs)offer a promising and cost-effective solution to boost network capacity,especially during emergencies and high-data-rate ***,integrating UAVs into the B5G networks presents new challenges,including resource scarcity,energy efficiency,resource allocation,optimal power transmission control,and maximizing overall *** paper presents a UAV-assisted B5G communication system where UAVs act as ABSs,and introduces the Deep Reinforcement Learning(DRL)based Energy Efficient Resource Allocation(Deep-EERA)*** efficient DRL-based Deep Deterministic Policy Gradient(DDPG)mechanism is introduced for optimal resource allocation with the twin goals of energy efficiency and average throughput *** proposed Deep-EERA method learns optimal policies to conserve energy and enhance throughput within the dynamic and complex UAV-empowered B5G *** extensive simulations,we validate the performance of the proposed approach,demonstrating that it outperforms other baseline methods in energy efficiency and throughput maximization.

关键词： Deep Reinforcement Learning(DRL) Unmanned Aerial Vehicles(UAVs) resource allocation energy efficiency 5G and beyond network

来源：评论

学校读者我要写书评

暂无评论

An Improved Dual-Path Attention Mechanism with Enhanced Discriminative Power for Asset Comparison Based on ResNet50 5

An Improved Dual-Path Attention Mechanism with Enhanced Disc...

引用

5th International Conference on Advances in Electrical, Electronics and Computing technology, EECT 2025

作者： Jing, Ying Zhang, Li School of Computer Science Beijing Information Science and Technology University Beijing China School of Management Science and Engineering Beijing Information Science and Technology University Beijing China

ISBN: (纸本)9798331541545

With the wide application of deep convolutional neural networks (CNNs), higher requirements are put forward for feature extraction and discriminant ability in asset comparison tasks. The rapid development of CNNs has significantly promoted the performance improvement of computer vision tasks, in which ResNet50, as the mainstream feature extraction network, effectively alleviates the problem of gradient disappearance in deep networks by means of residual learning. However, despite ResNet50's excellent performance in many visual tasks, there is still room for improvement in the effectiveness of feature expression and the mechanism of attention to key areas in response to changes in complex scenes. Therefore, this paper puts forth a new network architecture with improved dual-path attention mechanism and discriminative feature enhancement module based on ResNet50 (DADF-ResNet50). This method strengthens important features in channel and spatial dimensions respectively through the improved dual-path attention mechanism to improve the model's focusing ability on key asset areas. At the same time, the discriminating power enhancement module further optimizes the distinguishing ability of features and effectively improves the feature comparison performance between different assets. We design and implement the network architecture of this approach and evaluate it comprehensively in the asset comparison task. The experimental results show that the model has better distinguishing ability and feature extraction efficiency in diverse asset comparison scenarios, which is helpful to solve the asset identification problem in complex scenarios. © 2025 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Enabling Ultralow-Latency Services With Ubiquitous Mobility by Means of a Compact Network Architecture

引用

IEEE Transactions on Mobile Computing 2025年第6期24卷 4858-4873页

作者： Cai, Guiliang Wu, Qiang Wang, Ran Zhi, Lianyi Fu, Xiaoming Zhang, Hongke Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology Nanjing210016 China Faculty of Mathematics and Computer Science University of Göettingen Göttingen37073 Germany Beijing Jiaotong University School of Electronic and Information Engineering Beijing100044 China

With the rapid development of emerging services such as cellular vehicle-to-everything and immersive video service, network connections have further evolved from tangible physical connections to intangible virtual connections such as content, services, and computing resources, and the application scenarios have become more abundant. The mobile ultra-service, which is characterized by ultra-low latency, ultra-high reliability, and ubiquitous mobility, is becoming one of the most representative traffic types. However, the existing mobile network architecture has not evolved sufficiently to meet the specific requirements of these mobile ultra-services, the mobility anchors introduce unnecessary node and link latency, leaving space for further optimization. A compact network architecture (ComArch) is proposed in this paper for ultralow-latency services with ubiquitous mobility. ComArch is designed with a mapping control plane and a generalized forwarding plane to collaboratively implement packet forwarding in mobile scenarios. The generalized forwarding plane handles packet forwarding, while the mapping control plane manages terminals’ identifier and locator mapping entries. The node latency introduced by mobility anchors is eliminated, and an efficient routing scheme is proposed to find the optimal mandatory nodes in the forwarding path, thereby reducing unnecessary link latency. Experimental results show that ComArch can effectively reduce end-to-end delay while saving resources. © 2002-2012 IEEE.

关键词： Vehicle to Everything

来源：评论

学校读者我要写书评

暂无评论

AR-Light: Enabling Fast and Lightweight Multi-User Augmented Reality via Semantic Segmentation and Collaborative View Synchronization

引用

IEEE Transactions on computers 2025年第6期74卷 2073-2086页

作者： Wen, Yu Shah, Aamir Bader Cao, Ruizhi Zhang, Chen Chen, Jiefu Wu, Xuqing Xie, Chenhao Fu, Xin University of Houston Department of Electrical and Computer Engineering HoustonTX77204 United States Beihang University School of Computer Science and Engineering Beijing100191 China University of Houston Department of Information Science Technology HoustonTX77204 United States

Multi-user Augmented Reality (MuAR) allows multiple users to interact with shared virtual objects, facilitated by exchanging environment information. Current MuAR systems rely on 3D point clouds for real-world analysis, view synchronization, object rendering, and movement tracking. However, the complexity of 3D point clouds leads to significant processing delays, with approximately 80% of overhead in commercial frameworks. This hampers usability and degrades user experience. Our analysis reveals that maintaining the facing side of the real-world scene in a stable environment provides sufficient information for virtual object placement and rendering. To address this, we introduce a lightweight quadtree structure, representing 2D scenes through semantic segmentation and geometry, as an alternative to 3D point clouds. Additionally, we propose a novel correction method to handle potential shifts in virtual object placement during view synchronization among users. Combining all designs, we implement a fast and lightweight MuAR framework named AR-Light and test our framework on commercial AR devices. The evaluation results on real-world applications demonstrate that AR-Light can achieve high performance in various real-world scenes while maintaining a comparable virtual object placement accuracy. © 1968-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

DH-VTON: Deep Text-Driven Virtual Try-On via Hybrid Attention Learning

DH-VTON: Deep Text-Driven Virtual Try-On via Hybrid Attentio...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Wei, Jiabao Ma, Zhiyuan School of Computer Science and Technology Beijing Institute of Technology Beijing China Department of Eletronic Engineering Tsinghua University Beijing China

ISBN: (纸本)9798350368741

Virtual Try-ON (VTON) aims to synthesis specific person images dressed in given garments, which recently receives numerous attention in online shopping scenarios. Currently, the core challenges of the VTON task mainly lie in the fine-grained semantic extraction (i.e., deep semantics) of the given reference garments during depth estimation and effective texture preservation when the garments are synthesized and warped onto human body. To cope with these issues, we propose DH-VTON, a deep text-driven virtual try-on model featuring a special hybrid attention learning strategy and deep garment semantic preservation module. By standing on the shoulder of a well-built pre-trained paint-by-example (abbr. PBE) approach, we present our DH-VTON pipeline in this work. Specifically, to extract the deep semantics of the garments, we first introduce InternViT-6B as fine-grained feature learner, which can be trained to align with the large-scale intrinsic knowledge with deep text semantics (e.g., "neckline" or "girdle") to make up for the deficiency of the commonly adopted CLIP encoder. Based on this, to enhance the customized dressing abilities, we further introduce Garment-Feature ControlNet Plus (abbr. GFC+) module and propose to leverage a fresh hybrid attention strategy for training, which can adaptively integrate fine-grained characteristics of the garments into the different layers of the VTON model, so as to achieve multi-scale features preservation effects. Extensive experiments on several representative datasets demonstrate that our method outperforms previous diffusion-based and GAN-based approaches, showing competitive performance in preserving garment details and generating authentic human images. © 2025 IEEE.

关键词： Hybrid attention learning Stable diffusion Virtual try-on

来源：评论

学校读者我要写书评

暂无评论

Progressive Community Merging Cooperative Coevolution Algorithm for Influence Blocking Maximization in Social Networks

引用

IEEE Transactions on Network science and engineering 2025年第3期12卷 2093-2106页

作者： Gu, Ming Zhao, Tian-Fang Zhong, Jinghui Chen, Wei-Neng South China University of Technology School of Computer Science and Engineering Guangzhou510006 China South China University of Technology State Key Laboratory of Subtropical Building and Urban Science Guangzhou510006 China Jinan University School of Journalism and Communication Guangzhou510610 China

The widespread adoption of online social networks (OSNs) has facilitated social interaction and knowledge dissemination while raising concerns about extensive negative information propagation. Competitive propagation of positive and negative information can mitigate negative impacts. Influence blocking maximization (IBM) identifies a set of nodes initiating positive information propagation in OSNs to maximize blocking negative influence. This paper first introduces an effective influence blocking estimator (DPADV) that replaces computationally expensive Monte Carlo simulations. DPADV can calculate the approximate diffusion value for nodes activated by negative information in any hop neighborhood. Meanwhile, to address the challenges of complexity and computational efficiency brought about by the continuous expansion of OSNs, we propose the progressive community merging cooperative coevolution (PCMCC) algorithm. PCMCC divides the search space into communities and initializes a subpopulation for each community. Each subpopulation is responsible for optimizing a community, thereby implementing a divide-and-conquer approach. To enhance collaboration among communities and global exploration, we employed a progressive community merging strategy, supplemented by multi-population evolution strategies to guide the search towards global optima. Additionally, we developed an efficient heuristic metric for evaluating node importance, which is used to design population crossover and local search in the evolutionary scheme. Experimental results on six real-world and four synthetic networks demonstrate that PCMCC exhibits competitive performance compared to state-of-the-art algorithms, achieving near-greedy performance with lower time complexity. © 2013 IEEE.

关键词： Expectation maximization algorithm

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：