检索结果-内蒙古大学图书馆

Toward actionable testing of deep learning models

science China(Information sciences) 2023年第7期66卷 300-302页

作者： Yingfei XIONG Yongqiang TIAN Yepang LIU Shing-Chi CHEUNG Key Laboratory of High Confidence Software Technologies(Peking University) Ministry of Education School of Computer Science Peking University Department of Computer Science and Engineering The Hong Kong University of Science and Technology School of Computer Science University of Waterloo Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Department of Computer Science and Engineering Southern University of Science and Technology

Deep learning has become an important computational paradigm in our daily lives with a wide range of applications,from authentication using facial recognition to autonomous driving in smart vehicles. The quality of the deep learning models, i.e., neural architectures with parameters trained over a dataset, is crucial to our daily living and economy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SLRL: Semi-Supervised Local Community Detection Based on Reinforcement Learning 39

SLRL: Semi-Supervised Local Community Detection Based on Rei...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Ni, Li Ye, Rui Luo, Wenjian Zhang, Yiwen Zhang, Lei Sheng, Victor S. School of Computer Science and Technology Anhui University China Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies China School of Computer Science and Technology Harbin Institute of Technology China School of Department of Computer Science Texas Tech University United States

ISBN: (纸本)157735897X

Most existing semi-supervised community detection algorithms leverage known communities to learn community structures, subsequently identifying communities that align with these learned community structures. However, differences in community structures may render the community structures learned by these methods inappropriate for the community containing the given node of interest. As a result, the identified community may exclude the given node or be of poor quality. Inspired by the success of reinforcement learning, we propose a Semi-supervised Local community detection method based on Reinforcement Learning, named SLRL, which only explores parts of the network surrounding the given node. It first extracts the local structure around a given node with an extractor, followed by selecting communities that are similar to this local structure to distill useful communities. These selected communities are employed to train the expander, which expands the community containing a given node. Experimental results demonstrate that SLRL outperforms state-of-the-art algorithms on five real-world datasets. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A novel Locomotion Rule Rmbedding Long Short-Term Memory Network with Attention for Human Locomotor Intent Classification Using Multi-Sensors Signals

引用

computers, Materials & Continua 2024年第6期79卷 4349-4370页

作者： Jiajie Shen Yan Wang Dongxu Zhang Key Laboratory of Symbol Computation and Knowledge Engineering Ministry of EducationColleague of Computer Science and TechnologyJilin UniversityChangchun130012China College of Software and Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of EducationJilin UniversityChangchun130012China

Locomotor intent classification has become a research hotspot due to its importance to the development of assistive robotics and wearable *** work have achieved impressive performance in classifying steady locomotion ***,it remains challenging for these methods to attain high accuracy when facing transitions between steady locomotion *** to the similarities between the information of the transitions and their adjacent steady ***,most of these methods rely solely on data and overlook the objective laws between physical activities,resulting in lower accuracy,particularly when encountering complex locomotion modes such as *** address the existing deficiencies,we propose the locomotion rule embedding long short-term memory(LSTM)network with Attention(LREAL)for human locomotor intent classification,with a particular focus on transitions,using data from fewer sensors(two inertial measurement units and four goniometers).The LREAL network consists of two levels:One responsible for distinguishing between steady states and transitions,and the other for the accurate identification of locomotor *** classifier in these levels is composed of multiple-LSTM layers and an attention *** introduce real-world motion rules and apply constraints to the network,a prior knowledge was added to the network via a rule-modulating *** method was tested on the ENABL3S dataset,which contains continuous locomotion date for seven steady and twelve transitions *** results showed that the LREAL network could recognize locomotor intents with an average accuracy of 99.03%and 96.52%for the steady and transitions states,*** is worth noting that the LREAL network accuracy for transition-state recognition improved by 0.18%compared to other state-of-the-art network,while using data from fewer sensors.

关键词： Lower-limb prosthetics deep neural networks motion classification

来源：评论

学校读者我要写书评

暂无评论

A novel Quantization and Model Compression Approach for Hardware Accelerators in Edge Computing

引用

computers, Materials & Continua 2024年第8期80卷 3021-3045页

作者： Fangzhou He Ke Ding DingjiangYan Jie Li Jiajun Wang Mingzhe Chen State Key Laboratory of Intelligent Vehicle Safety Technology Chongqing401133China Foresight Technology Institute Chongqing Changan Automobile Co.Ltd.Chongqing400023China School of Computer Science and Engineering Chongqing University of Science and TechnologyChongqing401331China

Massive computational complexity and memory requirement of artificial intelligence models impede their deploy-ability on edge computing devices of the Internet of Things(IoT).While Power-of-Two(PoT)quantization is pro-posed to improve the efficiency for edge inference of Deep Neural Networks(DNNs),existing PoT schemes require a huge amount of bit-wise manipulation and have large memory overhead,and their efficiency is bounded by the bottleneck of computation latency and memory *** tackle this challenge,we present an efficient inference approach on the basis of PoT quantization and model *** integer-only scalar PoT quantization(IOS-PoT)is designed jointly with a distribution loss regularizer,wherein the regularizer minimizes quantization errors and training ***,two-stage model compression is developed to effectively reduce memory requirement,and alleviate bandwidth usage in communications of networked heterogenous learning *** product look-up table(P-LUT)inference scheme is leveraged to replace bit-shifting with only indexing and addition operations for achieving low-latency computation and implementing efficient edge ***,comprehensive experiments on Residual Networks(ResNets)and efficient architectures with Canadian Institute for Advanced Research(CIFAR),ImageNet,and Real-world Affective Faces Database(RAF-DB)datasets,indicate that our approach achieves 2×∼10×improvement in the reduction of both weight size and computation cost in comparison to state-of-the-art methods.A P-LUT accelerator prototype is implemented on the Xilinx KV260 Field Programmable Gate Array(FPGA)platform for accelerating convolution operations,with performance results showing that P-LUT reduces memory footprint by 1.45×,achieves more than 3×power efficiency and 2×resource efficiency,compared to the conventional bit-shifting scheme.

关键词： Edge computing model compression hardware accelerator power-of-two quantization

来源：评论

学校读者我要写书评

暂无评论

A novel Divide and Conquer Solution for Long-term Video Salient Object Detection

引用

Machine Intelligence Research 2024年第4期21卷 684-703页

作者： Yun-Xiao Li Cheng-Li-Zhao Chen Shuai Li Ai-Min Hao Hong Qin State Key Laboratory of Virtual Reality Technology and Systems Beihang UniversityBeijing 100191China College of Computer Science and Technology China University of Petroleum(East China)Qingdao 266580China Department of Computer Science Stony Brook UniversityNew York 11794USA

Recently,a new research trend in our video salient object detection(VSOD)research community has focused on enhancing the detection results via model self-fine-tuning using sparsely mined high-quality keyframes from the given *** such a learning scheme is generally effective,it has a critical limitation,i.e.,the model learned on sparse frames only possesses weak generalization *** situation could become worse on“long”videos since they tend to have intensive scene ***,in such videos,the keyframe information from a longer time span is less relevant to the previous,which could also cause learning conflict and deteriorate the model ***,the learning scheme is usually incapable of handling complex pattern *** solve this problem,we propose a divide-and-conquer framework,which can convert a complex problem domain into multiple simple ***,we devise a novel background consistency analysis(BCA)which effectively divides the mined frames into disjoint *** for each group,we assign an individual deep model on it to capture its key attribute during the fine-tuning *** the testing phase,we design a model-matching strategy,which could dynamically select the best-matched model from those fine-tuned ones to handle the given testing *** experiments show that our method can adapt severe background appearance variation coupling with object movement and obtain robust saliency detection compared with the previous scheme and the state-of-the-art methods.

关键词： Video salient object detection background consistency analysis weakly supervised learning long-term information background shift.

来源：评论

学校读者我要写书评

暂无评论

Real-Time DAG Task Allocation Strategy for Multiprocessor by Optimistic Parallelism 24

Real-Time DAG Task Allocation Strategy for Multiprocessor by...

引用

24th IEEE International Conference on Communication technology, ICCT 2024

作者： Chen, Yumeng Liu, Songlin Zhou, Jianhong Ling, Xiang University of Electronic Science and Technology of China National Key Laboratory of Science and Technology on Communications China Xihua University School of Computer and Software Engineering China

ISBN: (纸本)9798350363760

In modern real-time operating systems, complex task loads are often modeled as directed acyclic graphs (DAG) and executed in parallel on multiprocessor systems. The topological constraints present in DAG tasks prevent the DAG task allocation strategy from fully leveraging the parallel advantage of multiprocessor platforms. We analyze the parallelism of DAG tasks and propose a method to limit the number of processors that can be executed by DAG tasks. This method provides more idle processors and processor time for high parallelism DAG tasks by limiting the number of processors that can be executed by low parallelism DAG tasks. In addition, we develope optimistic parallelism scheduling algorithm (OPSA) based on the remaining utilization with low complexity. The experimental results show that our DAG task allocation strategy fully utilizes processor resources, reduces interference between DAG tasks, and achieves better worst case response time (WCRT) analysis results and shorter actual task's WCRT on real-time multiprocessor system. © 2024 IEEE.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence for Edge Service Optimization in Internet of Vehicles:A Survey

引用

Tsinghua science and technology 2022年第2期27卷 270-287页

作者： Xiaolong Xu Haoyuan Li Weijie Xu Zhongjian Liu Liang Yao Fei Dai the School of Computer and Software Nanjing University of Information Science and TechnologyNanjing 210044China the Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology(CICAEET) Nanjing University of Information Science and Technology and EngineeringNanjing 210044China the State Key Laboratory for Novel Software Technology Nanjing UniversityNanjing 210023China the School of Big Data and Intelligence Engineering Southwest Forestry UniversityKunming 650233China

The Internet of Vehicles(IoV)plays a crucial role in providing diversified services because of its powerful capability of collecting real-time ***,collected information is transmitted to a centralized resourceintensive cloud platform for service *** Computing(EC)that deploys physical resources near road-side units is involved in IoV to support real-time services for vehicular ***,many measures are adopted to optimize the performance of EC-enabled IoV,but they hardly help make dynamic decisions according to real-time *** Intelligence(AI)is capable of enhancing the learning capacity of edge devices and thus assists in allocating resources *** extensive research has employed AI to optimize EC performance,summaries with relative concepts or prospects are quite *** address this gap,we conduct an exhaustive survey about utilizing AI in edge service optimization in ***,we establish the general condition and relative concepts about IoV,EC,and ***,we review the edge service frameworks for IoV and explore the use of AI in edge server placement and service ***,we discuss a number of open issues in optimizing edge services with AI.

关键词： edge service internet of vehicles artificial intel igence

来源：评论

学校读者我要写书评

暂无评论

Super-resolution reconstruction of single image for latent features

引用

Computational Visual Media 2024年第6期10卷 1219-1239页

作者： Xin Wang Jing-Ke Yan Jing-Ye Cai Jian-Hua Deng Qin Qin Yao Cheng School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610000 China State Key Laboratory of Rail Transit Vehicle System Southwest Jiaotong UniversityChengdu 610000China School of Marine Engineering Guilin University of Electronic TechnologyGuilin 541000China School of Computer and Information Security Guilin University of Electronic TechnologyGuilin 541000China

Single-image super-resolution(SISR)typically focuses on restoring various degraded low-resolution(LR)images to a single high-resolution(HR)***,during SISR tasks,it is often challenging for models to simultaneously maintain high quality and rapid sampling while preserving diversity in details and texture *** challenge can lead to issues such as model collapse,lack of rich details and texture features in the reconstructed HR images,and excessive time consumption for model *** address these problems,this paper proposes a Latent Feature-oriented Diffusion Probability Model(LDDPM).First,we designed a conditional encoder capable of effectively encoding LR images,reducing the solution space for model image reconstruction and thereby improving the quality of the reconstructed *** then employed a normalized flow and multimodal adversarial training,learning from complex multimodal distributions,to model the denoising *** so boosts the generative modeling capabilities within a minimal number of sampling *** comparisons of our proposed model with existing SISR methods on mainstream datasets demonstrate that our model reconstructs more realistic HR images and achieves better performance on multiple evaluation metrics,providing a fresh perspective for tackling SISR tasks.

关键词： image superresolution reconstruction denoising diffusion probabilistic model normalized flow adversarial neural network variational auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Dynamic Pre-Warm Strategies for Reducing Cold-Start Latency in Serverless Edge Computing

Dynamic Pre-Warm Strategies for Reducing Cold-Start Latency ...

引用

International Conference on Mobile Ad-hoc and Sensor Networks, MSN

作者： Yicong Song Baoliu Ye Yue Zeng Xindong Wang Department of Computer Science and Technology State Key Laboratory for Novel Software Technology Nanjing University Nanjing University of Science and Technology

ISBN: (数字)9798331516024

ISBN: (纸本)9798331516031

Serverless edge computing uses lightweight containers to execute IoT applications on-demand, enhancing resource utilization efficiency at the network edge. However, cold-start latency, occurring when a container is instantiated from scratch, degrades IoT service responsiveness. Effective mitigation strategies are needed to maintain service quality. Current container caching strategies often fail to account for diverse workloads and predictable request patterns. This paper introduces a dynamic container pre-warm model that adjusts container states based on predicted future requests, mitigating cold-start overhead. Our approach employs a real-time algorithm for container instantiation and termination, balancing latency reduction and resource conservation. Inte-grating predictive analytics with adaptive resource management, we evaluate our model using a trace-driven simulation framework with real-world data. Our experiments show that the proposed strategy reduces cold starts by up to 60 % and improves average response times by 45 % compared to state-of-the-art methods. Additionally, our approach adapts to varying workloads, providing a scalable and robust solution for dynamic edge environments.

关键词： Adaptation models System performance Containers Predictive models Sensors Resource management Internet of Things Time factors Environmental management Edge computing

来源：评论

学校读者我要写书评

暂无评论

ICCG:low-cost and efficient consistency with adaptive synchronization for metadata replication

引用

Frontiers of computer science 2025年第1期19卷 53-70页

作者： Chenhao ZHANG Liang WANG Jing SHANG Zhiwen XIAO Limin XIAO Meng HAN Bing WEI Runnan SHEN Jinquan WANG State Key Laboratory of Software Development Environment Beihang UniversityBeijing 100191China School of Computer Science and Engineering Beihang UniversityBeijing 100191China School of Cyberspace Security Hainan UniversityHaikou 570228China China Mobile Information Technology Center Beijing 100033China

The rapid growth in the storage scale of wide-area distributed file systems (DFS) calls for fast and scalable metadata management. Metadata replication is the widely used technique for improving the performance and scalability of metadata management. Because of the POSIX requirement of file systems, many existing metadata management techniques utilize a costly design for the sake of metadata consistency, leading to unacceptable performance overhead. We propose a new metadata consistency maintenance method (ICCG), which includes an incremental consistency guaranteed directory tree synchronization (ICGDT) and a causal consistency guaranteed replica index synchronization (CCGRI), to ensure system performance without sacrificing metadata consistency. ICGDT uses a flexible consistency scheme based on the state of files and directories maintained through the conflict state tree to provide an incremental consistency for metadata, which satisfies both metadata consistency and performance requirements. CCGRI ensures low latency and consistent access to data by establishing a causal consistency for replica indexes through multi-version extent trees and logical time. Experimental results demonstrate the effectiveness of our methods. Compared with the strong consistency policies widely used in modern DFSes, our methods significantly improve the system performance. For example, in file creation, ICCG can improve the performance of directory tree operations by at least 36.4 times.

关键词： metadata management metadata replication consistency directory tree replica index

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：