检索结果-内蒙古大学图书馆

2024 IEEE International Conference on systems, Man, and Cybernetics, SMC 2024

作者： Wang, Bingshu Lin, Jia Zhuang, Xiaodong Zhang, Guanghui Chen, C.L. Philip School of Software Northwestern Polytechnical University Xi'an China Shenzhen University Guangdong Key Lab. of Intelligent Info. Processing and Shenzhen Key Laboratory of Media Security Shenzhen China School of Computer Science and Technology Shandong University Qingdao China School of Computer Science and Engineering South China University of Technology Guangzhou China

ISBN: (纸本)9781665410205

Concrete structural crack damage classification is of importance for road safety. This paper proposes a new method based on broad neural network for crack damage classification in concrete structures. It includes three stages. Firstly, a pre-trained deep neural network is used to extract the features from crack images. Secondly, principal component analysis is used to project the retrieved features from high dimensions to low dimensions. Thirdly, broad learning system is employed to predict the classification using the low-dimensional features. Experimental results demonstrate that this method reduces the model's training time and improves classification accuracy. © 2024 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

FedCrowd: Towards Promoting the Concurrency of the Server in Asynchronous Federated Learning 2

FedCrowd: Towards Promoting the Concurrency of the Server in...

引用

2nd International Symposium on Electrical, Electronics and Information Engineering, ISEEIE 2022

作者： Xi, Wen Liao, Yuying Zhou, Bin Wang, Haiyang Gao, Liqun College of Computer Science and Technology National University of Defence Technology Changsha China National University of Defence Technology Key Lab. of Software Engineering for Complex System Changsha China

ISBN: (数字)9781665468749

ISBN: (纸本)9781665468749

Federated learning is a privacy-preserving machine learning paradigm where a global model is trained collab.ratively across resource-limited devices such as mobile phones, wearable devices, and sensors while the training data remains on these devices. Recently proposed asynchronous federated learning framework FedAsync has addressed the problems occurred in synchronous federated learning such as wasting time waiting for all nodes to finish uploading or losing parts of the contribution of nodes due to some unexpected reasons. However, on the server-side, it is not suitable for concurrency due to the data race on the global model. We proposed a method named FedCrowd, where the server's concurrency is improved by elab.rately coordinating the working threads to relieve the data race on the global model. Several experiments are performed to show that our method could effectively increase concurrency of the server and reduce the training time. © 2022 IEEE.

关键词： Concurrent computing Training Wearable computers Training data Machine learning Collab.rative work Data models

来源：评论

学校读者我要写书评

暂无评论

A New Framework for Bounding Reachability Probabilities of Continuous-time Stochastic systems

arXiv

引用

arXiv 2023年

作者： Xue, Bai State Key Lab. of Computer Science Institute of Software Chinese Academy of Sciences Beijing China

This manuscript presents an innovative framework for constructing barrier functions to bound reachability probabilities for continuous-time stochastic systems described by stochastic differential equations (SDEs). The reachability probabilities considered in this paper encompass two aspects: the probability of reaching a set of specified states within a predefined finite time horizon, and the probability of reaching a set of specified states at a particular time instant. The barrier functions presented in this manuscript are developed either by relaxing a parabolic partial differential equation that characterizes the exact reachability probability or by applying the Grönwall's inequality. In comparison to the prevailing construction method, which relies on Doob's non-negative supermartingale inequality (or Ville's inequality), the proposed barrier functions provide stronger alternatives, complement existing methods, or fill gaps. Copyright © 2023, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

An Iterative Method for Computing Controlled Reach-Avoid Sets

An Iterative Method for Computing Controlled Reach-Avoid Set...

引用

American Control Conference (ACC)

作者： Dejin Ren Taoran Wu Bai Xue State Key Lab. of Computer Science CAS Institute of Software Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350382655

ISBN: (纸本)9798350382662

This paper focuses on addressing the problem of computing controlled reach-avoid sets for continuous-time systems modeled by polynomial ordinary differential equations with control inputs in order to ensure the safety and reachability of safety-critical systems. In a controlled reach-avoid set (CRAS), there exists a feedback controller capable of driving the closed-loop system into a target set of desired states while avoiding hazardous regions. The computation of a controlled reach-avoid set can be transformed into a search for a guidance-barrier function in existing literature. However, existing guidance-barrier functions tend to produce conservative CRASs due to certain flaws. To overcome this issue, a new guidance-barrier function is introduced, which generates less conservative CRASs compared to existing ones. Given the inherent nonlinearity associated with searching for guidance-barrier functions in the control setting, an iterative computational approach is proposed. This approach modifies controllers from previous iterations to find these functions and expand CRASs by solving convex optimizations. Finally, numerical results demonstrate the efficiency of our method in generating less conservative CRASs.

关键词： Three-dimensional displays Computational modeling Ordinary differential equations Polynomials Mathematical models Convex functions Safety

来源：评论

学校读者我要写书评

暂无评论

On Efficient Packet Batching and Resource Allocation for GPU based NFV Acceleration 31

On Efficient Packet Batching and Resource Allocation for GPU...

引用

31st IEEE/ACM International Symposium on Quality of Service, IWQoS 2023

作者： Zeng, Deze Zhu, Andong Gu, Lin Chen, Quan Guo, Minyi School of Computer Science China University of Geosciences Hubei Wuhan China Nanjing University State Key Lab. for Novel Software Technology Nanjing China School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Shanghai Jiao Tong University Department of Computer Science and Engineering Shanghai China

ISBN: (纸本)9798350399738

Network Function Virtualization (NFV) has already become an essential technology for improving the scalab.lity and flexibility of modern computer networks. The performance gap has become the main issue that impedes the development of NFV. GPUs, with massive parallel processors, are advocated to accelerate the Virtualized Network Functions (VNFs). However, the special architecture and workflow of GPUs introduce new challenges, especially on the batched processing, and resource allocation. In this paper, we propose GPU-based NFV Acceleration framework (GNFA) with an efficient packet batching and resource allocation solution. Considering the increased latency caused by the accumulation of the GPU kernel invoking overhead, we first invent a latency reduction mechanism called SM Performance Compensation (SPC). A Partition and Adjustment based Batching and Resource Allocation (PABARA) algorithm that jointly considers batch size tuning and GPU thread allocation is also proposed. We have practically implemented GNFA and extensively evaluated its performance on some well-known VNFs. The experiment results show that GNFA can effectively promote the GPU resource utilization and improve the NFV performance in terms of per-packet latency. © 2023 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Long- and Short- Term Sequential Recommendation with Temporal Interval 19

Long- and Short- Term Sequential Recommendation with Tempora...

引用

2022 IEEE SmartWorld, 19th IEEE International Conference on Ubiquitous Intelligence and Computing, 2022 IEEE International Conference on Autonomous and Trusted Vehicles Conference, 22nd IEEE International Conference on Scalab.e Computing and Communications, 2022 IEEE International Conference on Digital Twin, 8th IEEE International Conference on Privacy Computing and 2022 IEEE International Conference on Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PriComp/Metaverse 2022

作者： He, Kun Liu, Qiyan Li, Qianmu Meng, Shunmei Nanjing University of Science and Technology Nanjing China University of Toronto Computer Engineering Canada Nanjing University State Key Lab. for Novel Software Technology Nanjing China

ISBN: (纸本)9798350346558

Sequential recommendation aims to learn the changes of users' interests according to their historical behaviors and predict the most likely next item. Since user's historical behavior are sequential actions, user's interest in different time periods has different emphases. For predicting a user's next item, not only the recent behavior is important, but also long-term preference with all historical behaviors could not be ignored. Recently, self-attention based user preference modeling has drawn much attention for its advantages of fewer parameters and parallelism. However, most of the existing self-attention based model do not make a good distinction between the long-term and short-term preferences of users. Based on the observations, we design a sequential recommendation model based on a combination of long-term and short-term preference. On the one hand, considering the changes of users' interests in different periods, we divide the sequence of users' behaviors into different temporal windows. Then, we use GRU to capture users' interests in different temporal window. On the other hand, to better combine the global and local information of user's, we adopt a locally constrained multi-head attention mechanism based on Transformer encoder. On three real-world public datasets, we finally validate the efficacy of our method. © 2022 IEEE.

关键词： User profile

来源：评论

学校读者我要写书评

暂无评论

Adversarial Recovery Network for Low-Light Person Re-Identification 23

Adversarial Recovery Network for Low-Light Person Re-Identif...

引用

6th International Conference on Image and Graphics Processing, ICIGP 2023

作者： Xu, Liu Wang, Xiao Yuan, Xin Mao, Zhu Xu, Xin School of Computer Science and Technology Wuhan University of Science and Technology China School of Computer Science and Technology Wuhan University of Science and Technology ChinaHubei Prov. Key Lab. of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology China

ISBN: (纸本)9781450398572

Person re-identification (re-ID), which is to connect pedestrians with the same identity in different scenarios, has been more widely used in case detection of security. The re-ID datasets and technology have contributed greatly to advancing community research. There are currently no availab.e low-light datasets due to the difficulty of collecting and lab.ling images of people in low-light scenes. As a result, the vast majority of current re-ID methods do not have the recognition capability in low-light scenes. However, the recognition of persons in the low-light surveillance video has become an important issue to be solved for urban security, which plays an important role in improving the crime-solving rate of public safety departments and protecting people's lives and property. In this paper, we propose an Adversarial Recovery Network (ARN) that, unlike other approaches, enables a re-ID model with low-light recognition capability without the need for low-light data. Specifically, we first degrade the original images to generate multi-level low-light images;then obtain recovered images through the adversarial recovery module, and finally input the recovered images into the CNN network to extract pedestrian recognition knowledge. In addition, to validate the effectiveness of the proposed method, we also propose a low-light re-ID dataset (LOLR). Extensive experiments on LOLR have shown that our method outperforms other competitors by a wide margin. The dataset and code are availab.e online at https://***/xl02111/ARN. © 2023 ACM.

关键词： Recovery

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Study of Present Data Deduplication 23

A Comprehensive Study of Present Data Deduplication

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data science and systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Xue, Zilong Qian, Huixun Shen, Lingling Wu, Xiaotong School of Computer and Electronic Information Nanjing Normal University Nanjing China Nanjing University State Key Lab. for Novel Software Technology China

ISBN: (纸本)9781665494571

With the proliferation of the Internet of Things (IoT), various computing paradigms have been proposed and are developing rapidly these years, which led to the explosive increase of the data amount. However, the significant increase in data imposed a significant burden on contemporary server storage. Many approaches have been designed to tackle the challenge. Among them, deduplication is a quite effective data reduction technique, which received considerable attention from both academia and industry in the large-scale storage systems field. Data deduplication identifies redundant data at chunk level by using secure fingerprints, which not only removes replicated data, decreases the bandwidth, but also minimizes the storage usage and cost. This paper aims at describing the general framework of deduplication compression systems with duplicate and resemblance detection in detail. First, we use a flow chart to present the overall framework and process of the deduplication compression system. Then we summarize the existing algorithm applied to duplicate detection and resemblance detection. Additionally, we make a detailed evaluation of different resemblance detection algorithms. Finally, we make a conclusion of delta compressed prototype system, outline the open problems and shed a light on the future research directions in data deduplication systems. © 2021 IEEE.

关键词： Industries Costs Smart cities Prototypes Bandwidth Fingerprint recognition Explosives

来源：评论

学校读者我要写书评

暂无评论

Formally Verified C Code Generation from Hybrid Communicating Sequential Processes

Formally Verified C Code Generation from Hybrid Communicatin...

引用

IEEE/ACM International Conference on Cyber-Physical systems (ICCPS)

作者： Shuling Wang Zekun Ji Xiong Xu Bohua Zhan Qiang Gao Naijun Zhan State Key Lab. of Computer Science Institute of Software Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350369274

ISBN: (纸本)9798350369281

Hybrid Communicating Sequential Processes (HCSP) is a formal model for hybrid systems, including primitives for evolution along an ordinary differential equation (ODE), communication, and parallel composition. Code generation is needed to convert HCSP models into code that can be executed in practice, and the correctness of this conversion is essential to ensure that the generated code accurately reflects the formal model. In this paper, we propose a code generation algorithm from HCSP to C with POSIX library for concurrency. The main difficulties include how to bridge the gap between the synchronized communication model in HCSP and the use of mutexes for synchronization in C, and how to discretize evolution along ODEs and support interrupt of ODE evolution by communication. To prove the correctness of code generation, we define a formal semantics for POSIX C, and build transition system models for both HCSP and C programs. We then define an approximate bisimulation relation between traces of transition systems, and show that under certain robustness conditions for HCSP, the generated C program is approximately bisimilar to the original model. Finally, we evaluate the code generation algorithm on a detailed model for automatic cruise control, showing its utility on real-world examples.

关键词： Industries Codes Semantics Ordinary differential equations Mathematical models Robustness Hybrid power systems

来源：评论

学校读者我要写书评

暂无评论

HHLPar: Automated Theorem Prover for Parallel Hybrid Communicating Sequential Processes

arXiv

引用

arXiv 2024年

作者： Jin, Xiangyu Zhan, Bohua Wang, Shuling Zhan, Naijun Key Laboratory of System Software State Key Lab. of Computer Science ISCAS China University of Chinese Academy of Sciences China Huawei Technologies Co. Ltd. China National Key Laboratory of Space Integrated Information System ISCAS China School of Computer Science Peking University China

We present a tool called HHLPar for verifying hybrid systems modelled in Hybrid Communicating Sequential Processes (HCSP). HHLPar is built upon a Hybrid Hoare Logic for HCSP, which is able to reason about continuous-time properties of differential equations, as well as communication and parallel composition of parallel HCSP processes with the help of parameterised trace assertions and their synchronization. The logic was formalised and proved to be sound in Isabelle/HOL, which constitutes a trustworthy foundation for the verification conducted by HHLPar. HHLPar implements the Hybrid Hoare Logic in Python and supports automated verification: On one hand, it provides functions for symbolically decomposing HCSP processes, generating specifications for separate sequential processes and then composing them via synchronization to obtain the final specification for the whole parallel HCSP processes;On the other hand, it is integrated with external solvers for handling differential equations and real arithmetic properties. We have conducted experiments on a simplified cruise control system to validate the performance of the tool. Copyright © 2024, The Authors. All rights reserved.

关键词： Hybrid systems

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：