检索结果-内蒙古大学图书馆

Coarse-to-fine lightweight meta-embedding for ID-based recommendations

science China(Information sciences) 2025年第4期68卷 82-97页

作者： Yang WANG Haipeng LIU Zeqian YI Biao QIAN Meng WANG School of Computer Science and Information Engineering Hefei University of Technology College of Information and Intelligence Hunan Agricultural University

State-of-the-art recommender systems are increasingly focused on optimizing implementation efficiency, such as enabling on-device recommendations under memory constraints. Current methods commonly use lightweight embeddings for users and items or employ compact embeddings to enhance reusability and reduce memory usage. However, these approaches consider only the coarse-grained aspects of embeddings, overlooking subtle semantic nuances. This limitation results in an adversarial degradation of meta-embedding performance, impeding the system's ability to capture intricate relationships between users and items, leading to suboptimal recommendations. To address this, we propose a novel approach to efficiently learn meta-embeddings with varying grained and apply fine-grained meta-embeddings to strengthen the representation of their coarse-grained counterparts. Specifically, we introduce a recommender system based on a graph neural network, where each user and item is represented as a node. These nodes are directly connected to coarse-grained virtual nodes and indirectly linked to fine-grained virtual nodes, facilitating learning of multi-grained semantics. Fine-grained semantics are captured through sparse meta-embeddings, which dynamically balance embedding uniqueness and memory constraints. To ensure their sparseness, we rely on initialization methods such as sparse principal component analysis combined with a soft thresholding activation function. Moreover, we propose a weight-bridging update strategy that aligns coarse-grained meta-embedding with several fine-grained meta-embeddings based on the underlying semantic properties of users and items. Comprehensive experiments demonstrate that our method outperforms existing baselines. The code of our proposal is available at https://***/htyjers/C2F-MetaEmbed.

关键词： lightweight meta-embedding coarse-to-fine learning ID-based recommendations

来源：评论

学校读者我要写书评

暂无评论

TextFormer: A Query-based End-to-end Text Spotter with Mixed Supervision

引用

Machine Intelligence Research 2024年第4期21卷 704-717页

作者： Yukun Zhai Xiaoqiang Zhang Xiameng Qin Sanyuan Zhao Xingping Dong Jianbing Shen School of Computer Science Beijing Institute of TechnologyBeijing 100081China Department of Computer Vision Technology Baidu Inc.Beijing 100193China

End-to-end text spotting is a vital computer vision task that aims to integrate scene text detection and recognition into a unified *** methods heavily rely on region-of-interest(Rol)operations to extract local features and complex post-processing steps to produce final *** address these limitations,we propose TextFormer,a query-based end-to-end text spotter with a transformer ***,using query embedding per text instance,TextFormer builds upon an image encoder and a text decoder to learn a joint semantic understanding for multitask *** allows for mutual training and optimization of classification,segmentation and recognition branches,resulting in deeper feature sharing without sacrificing flexibility or ***,we design an adaptive global aggregation(AGG)module to transfer global features into sequential features for reading arbitrarilyshaped texts,which overcomes the suboptimization problem of Rol ***,potential corpus information is utilized from weak annotations to full labels through mixed supervision,further improving text detection and end-to-end text spotting *** experiments on various bilingual(i.e.,English and Chinese)benchmarks demonstrate the superiority of our *** on the TDA-ReCTS dataset,TextFormer surpasses the state-of-the-art method in terms of 1-NED by 13.2%.

关键词： End-to-end text spotting arbitrarily-shaped texts transformer mixed supervision multitask modeling.

来源：评论

学校读者我要写书评

暂无评论

Research on lightweight pavement disease detection model based on YOLOv7

引用

Journal of Intelligent and Fuzzy Systems 2024年第4期46卷 10573-10589页

作者： Wang, Chishe Li, Jun Wang, Jie Zhao, Weikang School of Computer Science and Engineering Anhui University of Science and Technology Huainan China Jinling Institute of Technology Nanjing China

Rapid urbanization has made road construction and maintenance imperative, but detecting road diseases has been time-consuming with limited accuracy. To overcome these challenges, we propose an efficient YOLOv7 road disease detection model. Our approach involves integrating MobilieNetV3 as the backbone feature extraction network to reduce the network's parameters and computational requirements. Additionally, we introduce the BRA attention module into the spatial pyramid pooling module to eliminate redundant information and enhance the network's feature representation capability. Moreover, we utilize the F-ReLU activation function in the backbone network, expanding the convolutional layers' receptive field range. To optimize the model's boundary loss, we employ the Wise-IoU loss function, which places more emphasis on the quality of ordinary samples and enhances the overall performance and generalization ability of the network. Experimental results demonstrate that our improved detection algorithm achieves a higher recall rate and mean average precision (mAP) on the public dataset (RDD) and the NJdata dataset in Nanjing's urban area. Specifically, compared to YOLOv7, our model increases the recall rate and mAP on RDD by 3.3% and 2.6%, respectively. On the NJdata dataset, our model improves the recall rate and mAP by 1.9% and 1.3%, respectively. Furthermore, our model reduces parameter and computational requirements by 30% and 22.5%, respectively, striking a balance between detection accuracy and speed. In conclusion, our road disease detection model presents an effective solution to address the challenges associated with road disease detection in urban areas. It offers improved accuracy, efficiency, and generalization capabilities compared to existing models. © 2024 - IOS Press. All rights reserved.

关键词： Roads and streets

来源：评论

学校读者我要写书评

暂无评论

Algorithms for online fault tolerance server consolidation

引用

Digital Communications and Networks 2025年第2期11卷 514-523页

作者： Boyu Li Bin Wu Meng Shen Hao Peng Weisheng Li Hong Zhang Jie Gan Zhihong Tian Guangquan Xu College of Intelligence and Computing Tianjin UniversityTianjin 300350China School of Cyberspace Science and Technology Beijing Institute of TechnologyBeijing 100081China School of Computer Science and Technology Zhejiang Normal UniversityJinhua 321004China School of Computer Science and Technology Chongqing University of Posts and TelecommunicationsChongqing 400054China School of Astronautics Beihang UniversityBeijing100191China School of Computer Science University of SydneySydneyNSW 20528Australia Cyberspace Institute of Advanced Technology Guangzhou UniversityGuangzhou 511370China

We study a novel replication mechanism to ensure service continuity against multiple simultaneous server *** this mechanism,each item represents a computing task and is replicated intoξ+1 servers for some integerξ≥1,with workloads specified by the amount of required *** one or more servers fail,the affected workloads can be redirected to other servers that host replicas associated with the same item,such that the service is not interrupted by the failure of up toξ*** requires that any feasible assignment algorithm must reserve some capacity in each server to accommodate the workload redirected from potential failed servers without overloading,and determining the optimal method for reserving capacity becomes a key *** existing algorithms that assume that no two servers share replicas of more than one item,we first formulate capacity reservation for a general arbitrary *** to the combinatorial nature of this problem,finding the optimal solution is *** this end,we propose a Generalized and Simple Calculating Reserved Capacity(GSCRC)algorithm,with a time complexity only related to the number of items packed in the *** conjunction with GSCRC,we propose a robust replica packing algorithm with capacity optimization(RobustPack),which aims to minimize the number of servers hosting replicas and tolerate multiple server *** theoretical analysis and experimental evaluations,we show that the RobustPack algorithm can achieve better performance.

关键词： Cloud computing Server consolidation Replica Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Real-time all-frequency global llumination with radiance caching

引用

Computational Visual Media 2024年第5期10卷 923-936页

作者： Youxin Xing Gaole Pan Xiang Chen Ji Wu Lu Wang Beibei Wang School of Software Shandong UniversityJinan 250101China School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210094China

Global illumination(GI)plays a crucial role in rendering realistic results for virtual exhibitions,such as virtual car *** scenarios usually include all-frequency bidirectional reflectance distribution functions(BRDFs),although their geometries and light configurations may be *** allfrequency BRDFs in real time remains challenging due to the complex light *** approaches,including precomputed radiance transfer,light probes,and the most recent path-tracing-based approaches(ReSTIR PT),cannot satisfy both quality and performance requirements ***,we propose a practical hybrid global illumination approach that combines ray tracing and cached GI by caching the incoming radiance with *** approach can produce results close to those of ofline renderers at the cost of only approximately 17 ms at runtime and is robust over all-frequency *** approach is designed for applications involving static lighting and geometries,such as virtual exhibitions.

关键词： real-time global illumination all-frequency BRDFs Haar wavelets radiance caching

来源：评论

学校读者我要写书评

暂无评论

Modeling and Analysis of OFDMA-NOMA-RA Protocol Considering Imperfect SIC in Multi-User Uplink WLANs

引用

computers, Materials & Continua 2024年第6期79卷 5273-5294页

作者： Hailing Yang Suoping Li Duo Peng School of Science Lanzhou University of TechnologyLanzhou730050China School of Computer&Communication Lanzhou University of TechnologyLanzhou730050China

To address the problems of network congestion and spectrum resources shortage in multi-user large-scale scenarios,this paper proposes a twice random access OFDMA-NOMA-RA protocol combining the advantages of orthogonal frequency division multiple access(OFDMA)and non-orthogonal multiple access(NOMA).The idea of this protocol is that OFMDA is used to divide the entire frequency field into multiple orthogonal resource units(RUs),and NOMA is used on each RU to enable more users to access the channel and improve spectrum *** on the protocol designed in this paper,in the case of imperfect successive interference cancellation(SIC),the probability of successful competition subchannels and the outage probability are derived for two scenarios:Users occupy the subchannel individually and users share the ***,when two users share the channel,the decoding order of the users and the corresponding probabilities are ***,the system throughput is *** achieve better outage performance in the system,the optimal power allocation algorithm is proposed in this paper,which enables the optimal power allocation strategy to be *** results show that the larger the imperfect SIC coefficient,the worse the outage performance of weak *** with pure OFDMA and NOMA,OFDMA-NOMA-RA always maintains an advantage when the imperfect SIC coefficient is less than a specific value.

关键词： Orthogonal frequency division multiple access(OFDMA) non-orthogonal multiple access(NOMA) random access(RA) imperfect successive interference cancellation(Imperfect SIC) outage probability

来源：评论

学校读者我要写书评

暂无评论

DedupChain: A secure Blockchain-enabled Storage System with Deduplication for Zero-Trust Network

引用

IEEE Journal on Selected Areas in Communications 2025年第6期43卷 2070-2086页

作者： Qi, Saiyu Wang, Qiuhao Wei, Wei Yang, Xu Zhao, Hongguang Liu, Yuhao Yang, Xu Qi, Yong Xi'an Jiaotong University School of Computer Science and Technology China Xi'an University of Technology School of Computer Science and Engineering China Alibaba Group China Northwestern Polytechnical University School of Electronics and Information China

Permissioned blockchain is a promising methodology to build zero-trust storage foundation with trusted data storage and sharing for the zero-trust network. However, the inherent full-backup feature of the permissioned blockchain poses potential data privacy risks and substantial storage costs, hindering its usage as a storage medium. These issues necessitate the usage of secure data deduplication technology to mitigate them. Unfortunately, current secure data deduplication schemes are predominantly designed with centralized cloud servers in mind and are not suitable for distributed blockchain systems. The reason is that the full backup feature of the permissioned blockchain renders a wide attack surface to offline brute-force and frequency analysis attacks. In response, we propose DedupChain, a secure blockchain-enabled storage system with deduplication for zero-trust networks. DedupChain employs a trusted execution environment (i.e., Inter SGX enclave) in conjunction with Oblivious RAM (ORAM) to offer a novel security guarantee named oblivious data deduplication, which empowers DedupChain with the ability to defend offline brute-force and frequency analysis attacks. DedupChain also proposes several novel techniques to address the security and efficiency issues raised by the SGX enclave. We implemented a system prototype of DedupChain and evaluated its performance metrics. Our experimental results show that DedupChain exhibits satisfactory operational delays, throughput, and storage overhead. Security analysis shows that DedupChain is robust enough to withstand several types of attacks. To the best of our knowledge, we are the first to apply secure data deduplication techniques to address data privacy and storage cost issues raised by permissioned blockchain when used as a zero-trust storage medium. © 1983-2012 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

AtomGAN:unsupervised deep learning for fast and accurate defect detection of 2D materials at the atomic scale

引用

science China(Information sciences) 2023年第6期66卷 102-114页

作者： Danpeng CHENG Wuxin SHA Zuo XU Shide LI Zhigao YIN Yuling LANG Shun TANG Yuan-Cheng CAO State Key Laboratory of Advanced Electromagnetic Engineering and Technology School of Electrical and Electronic EngineeringHuazhong University of Science and Technology School of Computer Science and Technology Huazhong University of Science and Technology CITIC Dicastal Co. Ltd.

The extraction of atomic-level material features from electron microscope images is crucial for studying structure-property relationships and discovering new materials. However, traditional electron microscope analyses rely on time-consuming and complex human operations; thus, they are only applicable to images with a small number of atoms. In addition, the analysis results vary due to observers' individual deviations. Although efforts to introduce automated methods have been performed previously, many of these methods lack sufficient labeled data or require various conditions in the detection process that can only be applied to the target material. Thus, in this study, we developed AtomGAN, which is a robust, unsupervised learning method, that segments defects in classical 2D material systems and the heterostructures of MoS2/WS2automatically. To solve the data scarcity problem, the proposed model is trained on unpaired simulated data that contain point and line defects for MoS2/WS2. The proposed AtomGAN was evaluated on both simulated and real electron microscope images. The results demonstrate that the segmented point defects and line defects are presented perfectly in the resulting figures, with a measurement precision of 96.9%. In addition, the cycled structure of AtomGAN can quickly generate a large number of simulated electron microscope images.

关键词： deep learning generative adversarial network defect detection atomic resolution 2D materials

来源：评论

学校读者我要写书评

暂无评论

Combining Innovative CVTNet and Regularization Loss for Robust Adversarial Defense

引用

Journal of computer science & technology 2024年第5期39卷 1078-1093页

作者： Wei-Dong Wang Zhi Li Li Zhang Laboratory of Public Big Data School of Computer Science and TechnologyGuizhou UniversityGuiyang 550025China

Deep neural networks(DNNs)are vulnerable to elaborately crafted and imperceptible adversarial *** the continuous development of adversarial attack methods,existing defense algorithms can no longer defend against them ***,numerous studies have shown that vision transformer(ViT)has stronger robustness and generalization performance than the convolutional neural network(CNN)in various ***,because the standard denoiser is subject to the error amplification effect,the prediction network cannot correctly classify all reconstruction ***,this paper proposes a defense network(CVTNet)that combines CNNs and ViTs that is appended in front of the prediction *** can effectively eliminate adversarial perturbations and maintain high ***,this paper proposes a regularization loss(L_(CPL)),which optimizes the CVTNet by computing different losses for the correct prediction set(CPS)and the wrong prediction set(WPS)of the reconstruction examples,*** evaluation results on several standard benchmark datasets show that CVTNet performs better robustness than other advanced *** with state-of-the-art algorithms,the proposed CVTNet defense improves the average accuracy of pixel-constrained attack examples generated on the CIFAR-10 dataset by 24.25%and spatially-constrained attack examples by 14.06%.Moreover,CVTNet shows excellent generalizability in cross-model protection.

关键词： deep learning adversarial defense vision transformer image reconstruction

来源：评论

学校读者我要写书评

暂无评论

A Random Fusion of Mix 3D and Polar Mix to Improve Semantic Segmentation Performance in 3D Lidar Point Cloud

引用

computer Modeling in Engineering & sciences 2024年第7期140卷 845-862页

作者： Bo Liu Li Feng Yufeng Chen School of Computer Science and Engineering Macao University of Science and TechnologyMacao999078China School of Computer Science and Artificial Intelligence Chaohu UniversityChaohu238000China Institute of Vehicle Information Control and Network Technology Hubei University of Automotive TechnologyShiyan442002China

This paper focuses on the effective utilization of data augmentation techniques for 3Dlidar point clouds to enhance the performance of neural network *** point clouds,which represent spatial information through a collection of 3D coordinates,have found wide-ranging *** augmentation has emerged as a potent solution to the challenges posed by limited labeled data and the need to enhance model generalization *** of the existing research is devoted to crafting novel data augmentation methods specifically for 3D lidar point ***,there has been a lack of focus on making the most of the numerous existing augmentation *** this deficiency,this research investigates the possibility of combining two fundamental data augmentation *** paper introduces PolarMix andMix3D,two commonly employed augmentation techniques,and presents a new approach,named *** of using a fixed or predetermined combination of augmentation methods,RandomFusion randomly chooses one method from a pool of options for each instance or *** innovative data augmentation technique randomly augments each point in the point cloud with either PolarMix or *** crux of this strategy is the random choice between PolarMix and Mix3Dfor the augmentation of each point within the point cloud data *** results of the experiments conducted validate the efficacy of the RandomFusion strategy in enhancing the performance of neural network models for 3D lidar point cloud semantic segmentation *** is achieved without compromising computational *** examining the potential of merging different augmentation techniques,the research contributes significantly to a more comprehensive understanding of how to utilize existing augmentation methods for 3D lidar point *** data augmentation technique offers a simple yet effective method to leverage the diversity of augmentation techniques and boost the ro

关键词： 3D lidar point cloud data augmentation RandomFusion semantic segmentation

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：