检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Yuwen, Cheng Wen, Guanghui Zhou, Jialing Luan, Meng Huang, Tingwen School of Automation Southeast University Nanjing210096 China State Key Laboratory of CNS/ATM Beijing Institute of Technology Beijing519088 China Department of System Science School of Mathematics Southeast University Nanjing211189 China Faculty of Computer Science and Control Engineering Shenzhen University of Advanced Technology Shenzhen218055 China

In this paper, a coalition game with local and coupling constraints is studied for uncertain Euler-Lagrange (EL) systems subject to disturbances with unknown bounds. In the coalition game, each agent collaborates with other agents within the same coalition to optimize its coalition’s cost function while simultaneously competing against agents in other coalitions. Under a distributed framework where each agent has access only to its own action, cost function, and constraint parameters, a distributed strategy is proposed to seek the Nash equilibrium (NE). By combining adaptive methods and sign functions, model uncertainties and disturbances with unknown bounds in the EL system are compensated and suppressed, respectively. Furthermore, an integration of adaptive methods and consensus protocols is employed to update the Lagrange multipliers of both local and coupling constraints. A dynamic average consensus is employed to estimate the gradient of coalition function, while the leader-following protocol is adopted to estimate the actions of other agents. By leveraging Lyapunov theory, the NE is proven to be asymptotically stable. Moreover, an unmanned surface vehicle swarm confrontation is meticulously modeled and analyzed in the coalition game framework. A diverse array of tasks, including formation, encirclement, and interception, are systematically formulated. A numerical example demonstrates the effectiveness of the proposed algorithm. Copyright © 2025, The Authors. All rights reserved.

关键词： Lagrange multipliers

来源：评论

学校读者我要写书评

暂无评论

Cross-Covariate Gait Recognition: A Benchmark

arXiv

引用

arXiv 2023年

作者： Zou, Shinan Fan, Chao Xiong, Jianbo Shen, Chuanfu Yu, Shiqi Tang, Jin School of Automation Central South University China Department of Computer Science and Engineering Southern University of Science and Technology China Research Institute of Trustworthy Autonomous System Southern University of Science and Technology China The University of Hong Kong Hong Kong

Gait datasets are essential for gait research. However, this paper observes that present benchmarks, whether conventional constrained or emerging real-world datasets, fall short regarding covariate diversity. To bridge this gap, we undertake an arduous 20-month effort to collect a cross-covariate gait recognition (CCGR) dataset. The CCGR dataset has 970 subjects and about 1.6 million sequences;almost every subject has 33 views and 53 different covariates. Compared to existing datasets, CCGR has both population and individual-level diversity. In addition, the views and covariates are well labeled, enabling the analysis of the effects of different factors. CCGR provides multiple types of gait data, including RGB, parsing, silhouette, and pose, offering researchers a comprehensive resource for exploration. In order to delve deeper into addressing cross-covariate gait recognition, we propose parsing-based gait recognition (ParsingGait) by utilizing the newly proposed parsing data. We have conducted extensive experiments. Our main results show: 1) Cross-covariate emerges as a pivotal challenge for practical applications of gait recognition. 2) ParsingGait demonstrates remarkable potential for further advancement. 3) Alarmingly, existing SOTA methods achieve less than 43% accuracy on the CCGR, highlighting the urgency of exploring cross-covariate gait recognition. Link: https://***/ShinanZou/CCGR. Copyright © 2023, The Authors. All rights reserved.

关键词： Gait analysis

来源：评论

学校读者我要写书评

暂无评论

Bmt-Enas: Branched Multi-Task Evolutionary Neural Architecture Search

SSRN

引用

SSRN 2024年

作者： Huang, Jianzun Zou, Juan Xia, Yizhang Hou, Zhanglu Liu, Yuan Ye, Songtao Lu, Qifeng Hunan Province ChinaHunan Engineering Research Center of Intelligent System Optimization and Security Xiangtan University Hunan Province Xiangtan411105 China Faculty of School of Computer Science School of Cyberspace Science of Xiangtan University Xiangtan411105 China Xi’an Jiaotong-Liverpool University Taicang Jiangsu Suzhou215400 China

Multi-task neural networks have achieved significant success in various field. However, developing a high-performance multi-task neural network typically necessitates a profound comprehension of multi-task learning principles and the complexities of the real-world problem at hand. In order to facilitate user-friendly multi-task network design, we propose Branched Multi-Task Evolutionary Neural Architecture Search (BMT-ENAS), based on evolutionary computation, to automatically design a network architecture. Specifically, we design a matrix that can be mapped to a branched multi-task architecture and adopt a genetic algorithm to evolve promising matrix individuals in combination with the proposed adaptive weight strategy. Compared with six stateof-the-art competitors, the experimental results illustrate that the BMT-ENAS not only outperforms existing automatic architecture design algorithms in terms of experimental effectiveness but also effectively balances the performance of each task © 2024, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Write-Optimized B^(+) Tree Index Technology for Persistent Memory

引用

Journal of computer science & Technology 2021年第5期36卷 1037-1050页

作者： Rui-Xiang Ma Fei Wu Bu-Rong Dong Meng Zhang Wei-Jun Li Chang-Sheng Xie Wuhan National Laboratory for Optoelectronics Huazhong University of Science and TechnologyWuhan 430070China Key Laboratory of Information Storage System Ministry of Education of ChinaWuhan 430070China Engineering Research Center of Data Storage Systems and Technology Huazhong University of Science and Technology Wuhan 430070China School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430070China Shenzhen DAPU Microelectronics Co. LtdShenzhen 518116China

Due to its low latency,byte-addressable,non-volatile,and high density,persistent memory(PM)is expected to be used to design a high-performance storage ***,PM also has disadvantages such as limited endurance,thereby proposing challenges to traditional index technologies such as B(+)tree.B(+)tree is originally designed for dynamic random access memory(DRAM)-based or disk-based systems and has a large write amplification *** high write amplification is detrimental to a PM-based *** paper proposes WO-tree,a write-optimized B(+)tree for ***-tree adopts an unordered write mechanism for the leaf nodes,and the unordered write mechanism can reduce a large number of write operations caused by maintaining the entry order in the leaf *** the leaf node is split,WO-tree performs the cache line flushing operation after all write operations are completed,which can reduce frequent data flushing ***-tree adopts a partial logging mechanism and it only writes the log for the leaf *** inner node recognizes the data inconsistency by the read operation and the data can be recovered using the leaf node information,thereby significantly reducing the logging ***,WO-tree adopts a lock-free search for inner nodes,which reduces the locking overhead for concurrency *** evaluate WO-tree using the Yahoo!Cloud Serving Benchmark(YCSB)*** with traditional B(+)tree,wB-tree,and Fast-Fair,the number of cache line flushes caused by WO-tree insertion operations is reduced by 84.7%,22.2%,and 30.8%,respectively,and the execution time is reduced by 84.3%,27.3%,and 44.7%,respectively.

关键词： persist memory B^(+)tree write amplication consistency YCSB(Yahoo!Cloud Serving Benchmark)

来源：评论

学校读者我要写书评

暂无评论

Stock Prediction Using Optimized LightGBM Based on Cost Awareness 5

Stock Prediction Using Optimized LightGBM Based on Cost Awar...

引用

5th IEEE International Conference on Cybernetics, CYBCONF 2021

作者： Zhao, Xiaosong Zhao, Qiangfu University of Aizu System Intelligence Laboratory Aizu Fukushima965-8580 Japan University of Aizu Graduate School of Computer Science and Engineering Aizu Fukushima965-8580 Japan

ISBN: (纸本)9781665403207

The application of machine learning in stock market prediction has generated widespread attention. This paper proposes a new method for short-term stock investment using optimized LightGBM based on cost awareness. By cost awareness here we define that the decision process is more aware of the false-positive errors, or is more aware of 'fake chances', and the cost for investment can be reduced. The main contribution of the research is to design an investment method for stock price prediction. Based on the principles of short-term investment and current quantitative investment research, we select a series of technical indicators that meet the requirements of this research to improve the reliability of the prediction results. We propose the concept of cost awareness to improve the accuracy of prediction. By comparing with the previous results, we demonstrated that the performance of the optimized model based on cost awareness could be improved significantly. we eventually examined and compared results with XGBoost and Random Forest, and found that LightGBM provides better performances in prediction accuracy, profitability, and risk control ability. © 2021 IEEE.

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

Weakly-Supervised Log-Based Anomaly Detection with Inexact Labels via Multi-Instance Learning

Weakly-Supervised Log-Based Anomaly Detection with Inexact L...

引用

International Conference on Software engineering (ICSE)

作者： Minghua He Tong Jia Chiming Duan Huaqian Cai Ying Li Gang Huang School of Software and Microelectronics Peking University Beijing China Institute for Artificial Intelligence Peking University Beijing China National Key Laboratory of Data Space Technology and System Beijing China School of Computer Science Peking University Beijing China National Engineering Research Center for Software Engineering Peking University Beijing China

ISBN: (数字)9798331505691

ISBN: (纸本)9798331505707

Log-based anomaly detection is essential for maintaining software availability. However, existing log-based anomaly detection approaches heavily rely on fine-grained exact labels of log entries which are very hard to obtain in real-world systems. This brings a key problem that anomaly detection models require supervision signals while labeled log entries are unavailable. Facing this problem, we propose a new labeling strategy called inexact labeling that instead of labeling an log entry, system experts can label a bag of log entries in a time span. Furthermore, we propose MIDLog, a weakly supervised log-based anomaly detection approach with inexact labels. We leverage the multiinstance learning paradigm to achieve explicit separation of anomalous log entries from the inexact labeled anomalous log set so as to deduce exact anomalous log labels from inexact labeled log sets. Extensive evaluation on three public datasets shows that our approach achieves an F1 score of over 85% with inexact labels.

关键词： Software Labeling Anomaly detection Software engineering

来源：评论

学校读者我要写书评

暂无评论

Enhancing Acceptance of Automated Vehicles Through Gamification: Insights from a Driving Simulator Study

SSRN

引用

SSRN 2025年

作者： Li, Chenchang Yin, Rao Yang, Bo Guan, Muhua Xie, Kun Nakano, Kimihiko Institute of Industrial Science University of Tokyo Tokyo153-8505 Japan School of Design & Innovation China Academy of Art Hangzhou311100 China School of Computer Science & System Engineering Kyushu Institute of Technology Fukuoka820-8502 Japan Department of Civil and Environmental Engineering Transportation Informatics Lab Old Dominion University NorfolkVA23529 United States

Automated vehicles hold great potential to revolutionize transportation, but subjective user acceptance remains a major obstacle to their widespread adoption and the realization of anticipated safety benefits. Recognizing this challenge, this study explores the role of gamification in enhancing user acceptance of automated vehicles and promoting safer driver engagement. By designing a gamified interaction framework and implementing it in a driving simulator experiment, the research addresses three core questions: Does gamification work? Why does it work? How can it be effectively implemented? In the experiment, participants experienced automated driving scenarios with integrated gamified interactions during overtaking tasks. Data on users' attitudes, behaviors, and eye movements were collected and analyzed through qualitative and quantitative methods. Furthermore, the study incorporated design research techniques, including interviews and user journey mapping, into the scientific framework to derive meaningful and actionable insights. The results reveal that gamification significantly improves user acceptance by enhancing Perceived Ease of Use and strengthening Behavioral Intentions. At a deeper mechanistic level, gamification enhances users' situational awareness and trust by increasing their perception efficiency during interactions and raising their risk tolerance, thereby shifting the balance between safety and reward. Based on these mechanisms and further findings, the study proposes three design strategies to maximize the benefits of gamification: designing system transparency, tailoring experiences to individual user characteristics, and leveraging incentive mechanisms to engage and motivate users effectively. This study's findings provide theoretical foundations and practical design directions for leveraging gamification to enhance human-vehicle interactions, utimately contributing to the development of safer and more efficient automated driving systems. © 2

关键词： Eye movements

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Architecture Search for Deep Graph Neural Networks

引用

IEEE Transactions on Big Data 2025年

作者： Dong, Yukang Pan, Fanxing Gui, Yi Jiang, Wenbin Wan, Yao Zheng, Ran Jin, Hai National Engineering Research Center for Big Data Technology Huazhong University of Science and Technology Wuhan430074 China Huazhong University of Science and Technology Service Computing Technology and System Laboratory Wuhan430074 China Huazhong University of Science and Technology Cluster and Grid Computing Laboratory Wuhan430074 China Huazhong University of Science and Technology School of Computer Science and Technology Wuhan430074 China Zhejiang Lab Hangzhou311121 China

In recent years, Neural Architecture Search (NAS) has emerged as a promising approach for automatically discovering superior model architectures for deep Graph Neural Networks (GNNs). Different methods have paid attention to different types of search spaces. However, due to the time-consuming nature of training deep GNNs, existing NAS methods often fail to explore diverse search spaces sufficiently, which constrains their effectiveness. To crack this hard nut, we propose CAS-DGNN, a novel comprehensive architecture search method for deep GNNs. It encompasses four kinds of search spaces that are the composition of aggregate and update operators, different types of aggregate operators, residual connections, and hyper-parameters. To meet the needs of such a complex situation, a phased and hybrid search strategy is proposed to accommodate the diverse characteristics of different search spaces. Specifically, we divide the search process into four phases, utilizing evolutionary algorithms and Bayesian optimization. Meanwhile, we design two distinct search methods for residual connections (All-connected search and Initial Residual search) to streamline the search space, which enhances the scalability of CAS-DGNN. The experimental results show that CAS-DGNN achieves higher accuracy with competitive search costs across ten public datasets compared to existing methods. © 2015 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Layered Stiffness Detection of Ballastless Track Based on Loading Force and Multiple Sinking Values

SSRN

引用

SSRN 2024年

作者： Miao, Shuaijie Gao, Liang Xin, Tao Yin, Hui Huang, Yonggui Xiao, Hong Cai, Xiaopei School of Civil Engineering Beijing Jiaotong University Beijing100044 China Frontiers Science Center for Smart High-speed Railway System Beijing Jiaotong University Beijing100044 China School of Computer and Information Technology Beijing Jiaotong University Beijing100044 China

Detecting the distribution of ballastless track layered stiffness and the stiffness anomalies caused by sublayer defects is attractive and challenging. Firstly, a data collection method for the ballastless track stiffness detection with loading force and multiple sinking values is proposed. Compared to the overall track stiffness detection by a single loading force and single sinking value, the proposed collection method can obtain richer track stiffness information, providing data support for the track layered stiffness detection, including the overall track stiffness and the slab upper and bottom stiffness. The principle verification test of a full-scale ballastless track has verified the numerical simulation feasibility of this data collection method. A dataset for detecting abnormal stiffness under fastener defects, mortar void, subgrade settlement, and defect overlap is constructed through simulation. Then, the stiffness detection dataset is fed into the BP neural network optimized by particle swarm optimization to predict the track layered stiffness, including overall track stiffness, slab upper stiffness, and slab bottom stiffness. Abnormal stiffness is located by the fluctuation of the track layered stiffness series. Finally, the layered stiffness series are processed by image to construct the image dataset, and the image dataset is fed into the Resnnet 18 model pre-trained by transfer learning to extract the features with intact stiffness and abnormal stiffness caused by sublayer defects. The results show that the 25 track stiffness abnormal cases are accurately identified, and the comprehensive accuracy is 94.63%. © 2024, The Authors. All rights reserved.

关键词： Stiffness

来源：评论

学校读者我要写书评

暂无评论

VulPin: Finer-Grained Slicing for Pinpointing Vulnerability in Binary Programs 23

VulPin: Finer-Grained Slicing for Pinpointing Vulnerability ...

引用

Proceedings of the 2023 7th International Conference on computer science and Artificial Intelligence

作者： Abdulrahman Hamman Adama Chukkol Senlin Luo Haruna Yunusa Abdulganiyu Abdu Yusuf Abulfatahi Mohammed Information and Electronics Department of Information and Electronics. Beijing Institute of Technology Beijing China China Information and Electronics Information System and Security & Countermeasures Experimental Center Beijing Institute of Technology Beijing China China Automation Science and Electrical Engineering School of Automation Science and Electrical Engineering Beihang University Beijing China China Computer Science School of Computer Science and Technology Beijing Institute of Technology Beijing China China Information Technology School of Management and Information Technology Modibbo Adama University Yola Nigeria Nigeria

ISBN: (纸本)9798400708688

Identifying vulnerabilities in binary code with precision is a critical area of study in the field of software security. Most existing techniques fail to pinpoint the vulnerability's exact location, due to the extensive labelled data requirement and lack of targeted approach to improve on the problematic code. In this paper, we introduce VulPin, a system specifically designed to pinpoint vulnerabilities in binary code. This framework uses a hierarchical attention network that analyze program slices along with a vulnerability position matrix. Experimental outcomes indicate that our framework not only outperforms existing approaches but also pinpoints vulnerabilities more rapidly with a high Accuracy and a very low rate of false positives.

关键词： Binary codes KEYWORDS Program slice Neural networks Vulnerability detection Vulnerability pinpoint

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：