检索结果-内蒙古大学图书馆

Multi-Stage Confidence-Guided Diffusion and Emotional Bidirectional Mamba for Robust Speech Emotion Recognition

IEEE Signal Processing Letters 2025年 32卷 2184-2188页

作者： Liu, Yang Chen, Xin Li, Yongwei Wang, Longbiao Zhao, Zhen Qingdao University of Science and Technology School of Information Science and Technology Qingdao266061 China Chinese Academy of Sciences CAS Key Laboratory of Behavioral Science Institute of Psychology Beijing100089 China Tianjin University Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin300072 China

Speech Emotion Recognition (SER) in noisy environments is challenging due to the overlap between emotional and noise-related signals. We propose a novel emotion-diffusion approach to enhance SER performance in noisy conditions that transfers emotional information from clean to noisy speech through a multi-stage diffusion process. First, Mel Frequency Cepstral Coefficients and their delta features are extracted to capture emotional dynamics. Additionally, an Emotional Bidirectional Mamba Encoder with a Multi-time View Bidirectional State Space Model is designed to capture temporal emotional patterns. Next, the Emotion-Transferring Diffusion Network (ETDN) applies confidence filtering to retain key emotional features, ensuring effective emotion transfer despite noise. Finally, the Confidence-Guided Mutual Learning strategy refines noisy features for the Classification Network (CN), while the CN supervises the ETDN to maintain label consistency. Experiments on the IEMOCAP dataset show significant improvements in weighted accuracy, unweighted accuracy, and weighted F1 score across various signal-to-noise ratios. © 1994-2012 IEEE.

关键词： Speech analysis

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Large-Scale AI Models for Intelligent Railway Transportation Systems:Hierarchical Poisoning Attacks and Defenses in Federated Learning

引用

Computer Modeling in Engineering & Sciences 2024年第11期141卷 1305-1325页

作者： Yongsheng Zhu Chong Liu Chunlei Chen Xiaoting Lyu Zheng Chen Bin Wang Fuqiang Hu Hanxi Li Jiao Dai Baigen Cai Wei Wang School of Automation and Intelligence Beijing Jiaotong UniversityBeijing100044China Institute of Computing Technologies China Academy of Railway Sciences Corporation LimitedBeijing100081China School of Computer Science and Technology Beijing Jiaotong UniversityBeijing100044China Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing Jiaotong UniversityBeijing100044China Institute of Infrastructure Inspection China Academy of Railway Sciences Corporation LimitedBeijing100081China Zhejiang Key Laboratory of Multi-Dimensional Perception Technology Application and CybersecurityHangzhou310053China

The development of Intelligent Railway Transportation Systems necessitates incorporating privacy-preserving mechanisms into AI models to protect sensitive information and enhance system *** learning offers a promising solution by allowing multiple clients to train models collaboratively without sharing private ***,despite its privacy benefits,federated learning systems are vulnerable to poisoning attacks,where adversaries alter local model parameters on compromised clients and send malicious updates to the server,potentially compromising the global model’s *** this study,we introduce PMM(Perturbation coefficient Multiplied by Maximum value),a new poisoning attack method that perturbs model updates layer by layer,demonstrating the threat of poisoning attacks faced by federated *** experiments across three distinct datasets have demonstrated PMM’s ability to significantly reduce the global model’s ***,we propose an effective defense method,namely CLBL(Cluster Layer By Layer).Experiment results on three datasets have confirmed CLBL’s effectiveness.

关键词： Privacy-preserving intelligent railway transportation system federated learning poisoning attacks defenses

来源：评论

学校读者我要写书评

暂无评论

Runtime reconfiguration of data services for dealing with out-of-range stream fluctuation in cloud-edge environments

引用

Digital Communications and Networks 2022年第6期8卷 1014-1026页

作者： Shouli Zhang Chen Liu Xiaohong Li Yanbo Han Division of Intelligence and Computing Tianjin UniversityTianjin300072China Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data North China University of TechnologyBeijing10009China Cloud Computing Research Center North China University of TechnologyBeijing10009China

The integration of cloud and IoT edge devices is of significance in reducing the latency of IoT stream data processing by moving services closer to the *** this connection,a key issue is to determine when and where services should be *** service deployment strategies used to be static based on the rules defined at the design ***,dynamically changing IoT environments bring about unexpected situations such as out-of-range stream fluctuation,where the static service deployment solutions are not *** this paper,we propose a dynamic service deployment mechanism based on the prediction of upcoming stream *** effectively predict upcoming workloads,we combine the online machine learning methods with an online optimization algorithm for service deployment.A simulation-based evaluation demonstrates that,compared with those state-of-the art approaches,the approach proposed in this paper has a lower latency of stream processing.

关键词： IoT stream processing Edge computing Out-of-Range stream fluctuation Dynamic service deployment

来源：评论

学校读者我要写书评

暂无评论

Auto-PIP: Real-time Identification of Critical Performance Inflection Points in Software Stress Testing 35

Auto-PIP: Real-time Identification of Critical Performance I...

引用

35th IEEE International Symposium on Software Reliability Engineering Workshops, ISSREW 2024

作者： Zhang, Shenglin Xiong, Xiao Li, Mengyao Sun, Yongqian Zhao, Yongxin Chen, Xia Deng, Bowen Pei, Dan Nankai University China Haihe Laboratory of Information Technology Application Innovation China Tianjin Key Laboratory of Software Experience and Human Computer Interaction China Huawei Cloud Tsinghua University Beijing National Research Center for Information Science and Technology China

ISBN: (纸本)9798350367041

Conducting stress testing is essential to ensure the stability and performance of software systems post-launch. Among its various aspects, the real-time identification of the "Smooth Load Area"(SLA) is crucial for establishing performance benchmarks, identifying bottlenecks, guiding optimization strategies, and strategically allocating resources. However, existing methods are incapable of real-time inflection point detection or require manual intervention. This paper proposes Auto-PIP, an automated identification framework for performance inflection points based on key performance indicators (KPIs) during performance stress tests. Auto-PIP integrates trend testing algorithms with unsupervised anomaly detection algorithms to identify optimal operating points and estimate maximum capacity. Auto-PIP has been deployed in Huawei cloud, and the evaluation results show that Auto-PIP demonstrated 100% accuracy in optimal inflection point detection, a 41.7% leap over baselines, and 83.9% accuracy in maximum point identification, a 10.7% gain. Additionally, we have released the dataset to the public to promote ongoing research. © 2024 IEEE.

关键词： Software testing

来源：评论

学校读者我要写书评

暂无评论

Towards practical data alignment in production federated learning

引用

Frontiers of Computer Science 2025年第1期19卷 167-169页

作者： Yexuan SHI Wei YU Yuanyuan ZHANG Chunbo XUE Yuxiang ZENG Zimu ZHOU Manxue GUO Lun XIN Wenjing NIE State Key Laboratory of Complex&Critical Software Environment and Advanced Innovation Center for Future Blockchain and Privacy Computing Beihang UniversityBeijing 100191China Zhongguancun Pan Connected Mobile Communication Technology Innovation and Application Research Institute Beijing 100088China China Mobile Research Institute Beijing 100053China School of Data Science City University of Hong KongHong Kong 999077China

1 Introduction Federated learning has emerged as a promising par-adigm for collaborative model training that facilitates cooperation among multiple parties while ensuring data privacy[1].Successful alignment of data across parties is crucial for effective federated learning[2].This alignment involves harmonizing heterogeneous data from different parties to identify shared data for joint model *** set intersection(PSI)is a technique that allows the alignment of common entities between parties without revealing additional ***,efficiently performing data alignment with PSI in federated learning[3],especially when dealing with highly unbalanced data,remains challenging due to the low efficiency.

关键词： alignment dealing parties

来源：评论

学校读者我要写书评

暂无评论

Extracting Variable-Depth Logical Document Hierarchy from Long Documents:Method,Evaluation,and application

引用

Journal of Computer Science & technology 2022年第3期37卷 699-718页

作者： Rong-Yu Cao Yi-Xuan Cao Gan-Bin Zhou Ping Luo Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences Institute of Computing TechnologyChinese Academy of SciencesBeijing 100190China University of Chinese Academy of Sciences Beijing 100049China WeChat Search Application Department Tencent Holdings Ltd.Beijing 100080China Peng Cheng Laboratory Shenzhen 518066China

In this paper,we study the problem of extracting variable-depth"logical document hierarchy"from long documents,namely organizing the recognized"physical document objects"into hierarchical *** discovery of logical document hierarchy is the vital step to support many downstream applications(e.g.,passage-based retrieval and high-quality information extraction).However,long documents,containing hundreds or even thousands of pages and a variable-depth hierarchy,challenge the existing *** address these challenges,we develop a framework,namely Hierarchy Extraction from Long Document(HELD),where we"sequentially"insert each physical object at the proper position on the current *** whether each possible position is proper or not can be formulated as a binary classification *** further improve its effectiveness and efficiency,we study the design variants in HELD,including traversal orders of the insertion positions,heading extraction explicitly or implicitly,tolerance to insertion errors in predecessor steps,and so *** for evaluations,we find that previous studies ignore the error that the depth of a node is correct while its path to the root is *** such mistakes may worsen the downstream applications seriously,a new measure is developed for a more careful *** empirical experiments based on thousands of long documents from Chinese financial market,English financial market and English scientific publication show that the HELD model with the"root-to-leaf"traversal order and explicit heading extraction is the best choice to achieve the tradeoff between effectiveness and efficiency with the accuracy of 0.972,6,0.729,1 and 0.957,8 in the Chinese financial,English financial and arXiv datasets,***,we show that the logical document hierarchy can be employed to significantly improve the performance of the downstream passage retrieval *** summary,we conduct a systematic study on this task in terms of methods

关键词： logical document hierarchy long documents passage retrieval

来源：评论

学校读者我要写书评

暂无评论

A Failure Prediction Approach Supporting Multi Granularity Data Fusion for Large-scale cloud Storage Systems 22

A Failure Prediction Approach Supporting Multi Granularity D...

引用

5th International Conference on Computer Science and Software Engineering, CSSE 2022

作者： Cheng, Yongyang Zhang, Tao Luo, Jing China Telecom Cloud Computing Corporation Beijing China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781450397780

With the development of cloud computing and cloud storage technology, the data scale has grown rapidly. In order to store and process large-scale data, there are thousands of nodes and devices in the cloud storage center, resulting in a surge in the frequency of failures. In various types of failure events, storage device failure is the most important one. However, most cloud storage systems lack disk failure prediction mechanisms and could only replace disks after disk failures. It is particularly important to predict the potential risks in the system operation environment. In this paper, we propose a disk failure prediction approach that supports multi granularity data fusion, which solves problems of unbalanced samples, single data source, cross scenario model migration and insufficient generalization ability of prediction models in disk failure prediction. Through our proposed approach, the cloud storage system could accurately predict disk failures and actively push prediction results to users, so as to improve the pertinence and planning of the operation and maintenance work. The approach presented in this paper has been validated to be valid through a series of qualitative and quantitative experiments. © 2022 ACM.

关键词： Data fusion

来源：评论

学校读者我要写书评

暂无评论

Analysis of Factors Influencing Driver Lane-Changing Intentions Based on a Naturalistic Trajectory Data Set on Highways 24

Analysis of Factors Influencing Driver Lane-Changing Intenti...

引用

24th COTA International Conference of Transportation Professionals: Resilient, Intelligent, Connected, and Lowcarbon Multimodal Transportation, CICTP 2024

作者： Wang, Guangchen Lu, Guangquan Wang, Jinghua Liu, Miaomiao School of Transportation Science and Engineering Beihang Univ. Beijing China Beijing Key Laboratory for Cooperative Vehicle Infrastructure Systems and Safety Control National Engineering Laboratory for Comprehensive Transportation Big Data Application Technology Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang Univ. Beijing China

ISBN: (数字)9780784485484

ISBN: (纸本)9780784485484

Accurately analyzing and predicting driver lane-changing intentions is of paramount importance, as it significantly enhances the safety of self-driving vehicles in their decision-making processes, holding great promise for the future of intelligent transportation systems. This study extracts fundamental lane-changing characteristic parameters from the Next Generation Simulation (NGSIM) trajectory data set and employs Structural Equation Modeling (SEM) to construct a model that elucidates the factors influencing drivers’ lane-changing intentions. Furthermore, it provides insight into how lane-changing intentions are influenced by lane-changing behavior and explores the intricate connection between intention and associated risks. The results emphasize the substantial negative impact of the longitudinal speed difference between the self-vehicle and the preceding vehicle on lane-changing intentions. Additionally, the consideration of lane-changing risk plays a pivotal role in drivers’ decisions to engage in lane-changing behavior. These findings have significant implications for a deeper comprehension of lane-changing intentions, guiding lane-changing decisions, and enhancing overall traffic safety. © ASCE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

HTDet:A Clustering Method Using Information Entropy for Hardware Trojan Detection

引用

Tsinghua Science and technology 2021年第1期26卷 48-61页

作者： Renjie Lu Haihua Shen Zhihua Feng Huawei Li Wei Zhao Xiaowei Li the University of Chinese Academy of Sciences Beijing 101408China Beijing Institute of Computer Technology and Application Beijing 100854China the State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of SciencesBeijing 100190China

Hardware Trojans(HTs)have drawn increasing attention in both academia and industry because of their significant potential *** this paper,we propose HTDet,a novel HT detection method using information entropybased *** maintain high concealment,HTs are usually inserted in the regions with low controllability and low observability,which will result in that Trojan logics have extremely low transitions during the *** implies that the regions with the low transitions will provide much more abundant and more important information for HT *** HTDet applies information theory technology and a density-based clustering algorithm called Density-Based Spatial Clustering of applications with Noise(DBSCAN)to detect all suspicious Trojan logics in the circuit under *** DBSCAN is an unsupervised learning algorithm,that can improve the applicability of *** addition,we develop a heuristic test pattern generation method using mutual information to increase the transitions of suspicious Trojan *** on circuit benchmarks demonstrate the effectiveness of HTDet.

关键词： Hardware Trojan(HT)detection information entropy Density-Based Spatial Clustering of applications with Noise(DBSCAN) unsupervised learning clustering mutual information test patterns generation

来源：评论

学校读者我要写书评

暂无评论

An anomaly detection approach based on hybrid differential evolution and K-means clustering in crowd intelligence

引用

International Journal of Crowd Science 2021年第2期5卷 129-142页

作者： Liu, Jianran Liang, Bing Ji, Wen Institute of Computing Technology Chinese Academy of Sciences Beijing China Institute of Computing Technology Chinese Academy of Sciences Beijing China Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of Sciences Beijing China Fujian Key Laboratory of Granular Computing and Application Minnan Normal University Zhangzhou China University of Chinese Academy of Sciences Beijing China

Purpose: Artificial intelligence is gradually penetrating into human society. In the network era, the interaction between human and artificial intelligence, even between artificial intelligence, becomes more and more complex. Therefore, it is necessary to describe and intervene the evolution of crowd intelligence network dynamically. This paper aims to detect the abnormal agents at the early stage of intelligent evolution. Design/methodology/approach: In this paper, differential evolution (DE) and K-means clustering are used to detect the crowd intelligence with abnormal evolutionary trend. Findings: This study abstracts the evolution process of crowd intelligence into the solution process of DE and use K-means clustering to identify individuals who are not conducive to evolution in the early stage of intelligent evolution. Practical implications: Experiments show that the method we proposed are able to find out individual intelligence without evolutionary trend as early as possible, even in the complex crowd intelligent interactive environment of practical application. As a result, it can avoid the waste of time and computing resources. Originality/value: In this paper, DE and K-means clustering are combined to analyze the evolution of crowd intelligent interaction. © 2020, Jianran Liu, Bing Liang and Wen Ji.

关键词： K-means clustering

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：