检索结果-内蒙古大学图书馆

2024 IEEE International Conference on Robotics and Biomimetics, ROBIO 2024

作者： Li, Mingyi Li, Jiayin Liu, Hui Han, Lijin Liao, Dengting Zhang, Zekun Gao, Zong Liu, Baoshuai Jin, Zhongfeng School of Mechanical Engineering Beijing Institute of Technology Beijing100081 China School of Computing and Data Science Xiamen University Malaysia Selangor 43900 Malaysia Shanghai Zhangjiang Institute of Mathematicas Zhangjiang Institute of Shanghai Shanghai201203 China China North Vehicle Research Institute General Technology Department 100072 China Coordination Center of China GNational Computer Network Emergency Response Technical Team 100083 China

ISBN: (纸本)9781665481090

In light of the challenges associated with enhancing multimodal knowledge graph retrieval and improving embodied intelligence in dynamic environments, we propose the Multimodal Knowledge Graph Vision-Language System (MKGVL). This system is designed to enhance the reasoning and feedback capabilities of embodied intelligent systems by continuously updating the knowledge graph through real-time feedback from the Vision-Language Model (VLM). By integrating visual encoders, language models, and knowledge graph networks, MKGVL constructs a unified multimodal representation that enables adaptive decision-making and responsiveness to environmental changes. Experimental results show that MKGVL outperforms existing models in fine-grained retrieval tasks, achieving an 11.5% improvement in Rank1 accuracy and a mean Average Precision (mAP) of 97.49%. Further evaluations conducted on datasets such as ARKitScenes, MultiScan, and 3RScan highlight the model's robustness and adaptability. Additionally, MKGVL's deployment on embedded platforms like Jetson Orin demonstrates its efficiency in real-time multimodal tasks, particularly in resource-constrained environments. These findings underscore MKGVL's ability to deliver accurate and efficient multimodal processing, making it a strong solution for adaptive knowledge-based systems in complex settings. © 2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Resolving Spurious Temporal Location Dependency for Video Corpus Moment Retrieval

Resolving Spurious Temporal Location Dependency for Video Co...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Yishuo Zhang Lanshan Zhang Zhizhen Zhang Ziyi Wang Xiaohui Xie Wendong Wang School of Computer Science (National Pilot Software Engineering School) Beijing University of Posts and Telecommunications Beijing China Beijing Key Laboratory of Network System and Network Culture Beijing University of Posts and Telecommunications Beijing China Shenzhen International Graduate School Tsinghua University Shenzhen China Department of Computer Science and Technology Tsinghua University Beijing China

ISBN: (数字)9781665410205

ISBN: (纸本)9781665410212

Video Corpus Moment Retrieval aims to retrieve the relevant video from a large corpus and localize the corresponding moment within the target video based on a specific query. Existing methods have achieved promising accuracy and efficiency through dedicated model designs. However, we suggest these methods overly exploit dataset biases instead of semantics. This will lead to exaggerated performances on biased datasets and implicit a significant deficiency in generalizability, which is an important metric yet not considered in existing studies. In this paper, we observe the degradation caused by a spurious dependency and design a model to mitigate this harm. Specifically, we generate an Out-Of-Distributed (OOD) test set from a widely used TV Retrieval dataset, revealing the existing models' erroneous dependency on the temporal locations of target moments. Therefore, we utilize a theoretical Structural Causal Model (SCM) to dig into the roots of this dependency by constructing causal paths for the models. Furthermore, we propose a concrete Clip Location Deconfounding Model (CLDM) to disentangle the confounded video features into the content part and the location confounder part, then produce results with causal intervention. Experiments show that CLDM significantly alleviates the impact brought by dataset biases thus providing advanced generalizability among existing works.

关键词： Measurement Degradation TV Accuracy Semantics Cybernetics

来源：评论

学校读者我要写书评

暂无评论

Quadratically Constrained Myopic Adversarial Channels

Quadratically Constrained Myopic Adversarial Channels

引用

作者： Zhang, Yihan Vatedka, Shashank Jaggi, Sidharth Sarwate, Anand D. Department of Information Engineering Chinese University of Hong Kong Hong Kong Technion-Israel Institute of Technology Henry and Marilyn Taub Faculty of Computer Science Haifa3200003 Israel Institute of Science and Technology Austria Klosterneuburg3400 Austria Department of Communications and Electronics Palaiseau91120 France Institute of Network Coding Chinese University of Hong Kong Hong Kong Indian Institute of Technology Hyderabad Department of Electrical Engineering Hyderabad502285 India Department of Electrical Engineering Indian Institute of Technology Hyderabad Hyderabad502285 India University of Bristol School of Mathematics BristolBS8 1TH United Kingdom State University of New Jersey Rutgers Department of Electrical and Computer Engineering New BrunswickNJ08854 United States

We study communication in the presence of a jamming adversary where quadratic power constraints are imposed on the transmitter and the jammer. The jamming signal is allowed to be a function of the codebook, and a noncausal but noisy observation of the transmitted codeword. For a certain range of the noise-to-signal ratios (NSRs) of the transmitter and the jammer, we are able to characterize the capacity of this channel under deterministic encoding or stochastic encoding, i.e., with no common randomness between the encoder/decoder pair. For the remaining NSR regimes, we determine the capacity under the assumption of a small amount of common randomness (at most 2\log (n) bits in one sub-regime, and at most \Omega ({n}) bits in the other sub-regime) available to the encoder-decoder pair. Our proof techniques involve a novel myopic list-decoding result for achievability, and a Plotkin-type push attack for the converse in a subregion of the NSRs, both of which may be of independent interest. We also give bounds on the strong secrecy capacity of this channel assuming that the jammer is simultaneously eavesdropping. © 1963-2012 IEEE.

关键词： Transmitters

来源：评论

学校读者我要写书评

暂无评论

An Android Application for Unwanted Vehicle Detection and Counting

An Android Application for Unwanted Vehicle Detection and Co...

引用

Human-Centered Cognitive Systems (HCCS)

作者： Maeeda Khalid Adnan Ashraf Waqas Haider Bangyal Muddesar Iqbal School of Computer Science & Technology Beijing Institute of Technology (BIT) Beijing China Department of Computer Science Facuty of Computing Kohsar University Murree Murree Pakistan Department of Communication & Network Engineering Prince Sultan University Riyadh Saudi Arabia

The significance of vehicle identification and monitoring is increasing in the field of traffic management. The Intelligent Transportation System (ITS) is a highly efficient way to address the issue of traffic congestion in metropolitan areas and is a prominent focus in the development of smart cities. One specific application is the monitoring and forecasting of fluid flow. The suggested system aims to implement automated vehicle identification and recognition processing utilizing static image datasets. Significant progress in vehicle detection technology has been made due to the emergence of unmanned driving and intelligent transportation research. The suggested system utilizes the deep learning technique to investigate the vehicle detection algorithm, specifically employing the fundamental phase target detection algorithm known as the YOLO algorithm. Hence, the initial approach involves the manipulation of visual data from a publicly available collection of road vehicles for the purpose of training. A vehicle detection model is developed using the YOLO algorithm to demonstrate the detection performance separately. The suggested system’s contribution is in the enhancement of the conventional YOLO network’s architecture, enabling precise identification of vehicle targets.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning for Semantic Knowledge Base-Guided Online Feature Transmission in Dynamic Channels

Learning for Semantic Knowledge Base-Guided Online Feature T...

引用

IEEE International Conference on Communications (ICC)

作者： Xiangyu Gao Yaping Sun Dongyu Wei Xiaodong Xu Hao Chen Hao Yin Shuguang Cui Department of Electrical and Computer Engineering University of Washington Seattle USA Department of Broadband Communication Peng Cheng Laboratory Shenzhen China Future Network of Intelligent Institute (FNii) the Chinese University of Hong Kong (Shenzhen) Shenzhen China Beijing University of Posts and Telecommunications Beijing China School of Science and Engineering (SSE) and the Future Network of Intelligent Institute (FNii) the Chinese University of Hong Kong (Shenzhen) Shenzhen China

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

With the proliferation of edge computing, efficient AI inference on edge devices has become essential for intelligent applications such as autonomous vehicles and VR/AR. In this context, we address the problem of efficient remote object recognition by optimizing feature transmission between mobile devices and edge servers. We propose an online optimization framework to address the challenge of dynamic channel conditions and device mobility in an end-to-end communication system. Our approach builds upon existing methods by leveraging a semantic knowledge base to drive multi-level feature transmission, accounting for temporal factors and dynamic elements throughout the transmission process. To solve the online optimization problem, we design a novel soft actor-critic-based deep reinforcement learning system with a carefully designed reward function for real-time decision-making, overcoming the optimization difficulty of the NP-hard problem and achieving the minimization of semantic loss while respecting latency constraints. Numerical results showcase the superiority of our approach compared to traditional greedy methods under various system setups.

关键词： System performance Simulation Semantics Knowledge based systems Mobile handsets Real-time systems Object recognition

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised Medical Image Segmentation via Feature-perturbed Consistency

Semi-supervised Medical Image Segmentation via Feature-pertu...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Yang Yang Ruixuan Wang Tong Zhang Jingyong Su School of Computer Science and Technology Harbin Institute of Technology Shenzhen China Department of Network Intelligence Peng Cheng Laboratory Shenzhen China School of Computer Science and Engineering Sun Yat-sen University Guangzhou China

Although deep convolutional neural networks have achieved satisfactory performance in many medical image segmentation tasks, a considerable annotation challenge still needs to be solved, which is expensive and time-consuming for radiologists. Most existing popular semi-supervised methods mainly impose data-level perturbations (e.g., rotation, noising) or feature-level perturbations (e.g., MC dropout) on unlabeled data. In this paper, we propose a novel semi-supervised segmentation strategy with meaningful perturbations at the feature level to leverage abundant useful information naturally embedded in the unlabeled data. Specifically, we develop a dual-task network where the segmentation head produces multiple predictions with a perturbation module, and the reconstruction head further utilizes the semantic information to enhance segmentation performance. The proposed framework subtly perturbs the network at the feature-level to generate predictions which should be similar and consistent. However, enforcing them roughly to be consistent at all pixels harms stable training and neglects much delicate information. To better utilize those predictions and estimate the uncertainty, we further propose feature-perturbed consistency to exploit reliable regions for our framework to learn from. Extensive experiments on the public BraTS2020 dataset and the 2017 ACDC dataset confirm the efficiency and effectiveness of our method. In particular, the proposed method demonstrates remarkable superiority in the segmentation of boundary regions. The project is available at https://***/youngyzzZ/SFPC.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Random Forest Stock Prediction Model Based on Bayesian Optimization

A Random Forest Stock Prediction Model Based on Bayesian Opt...

引用

International Conference on Artificial Intelligence and Big Data (ICAIBD)

作者： Yajuan Zhang Xiuyan Zheng Sihan Yang Shangyu Meng Z. Y. Yang Xinghui Fei College of Information Engineering Hainan Vocational University of Science and Technology Haikou Hainan China School of Information and Software Engineering University of Electronic Science and Technology of China Sichuan Chengdu China Network engineer department Guiyang Shengteng Network Technology Co. LTD Guiyang China Electrical and Computer Engineer Deparment Colorado State University Fort Collins U.S.

ISBN: (数字)9798350385106

ISBN: (纸本)9798350385113

Random forest models often rely on human-determined parameters, which may introduce subjectivity and affect the degree of optimization of the model. To address this issue, we use a random forest approach based on Bayesian optimization to reduce the influence of human factors and improve the fitting. Bayesian optimization prevents the model from falling into the dilemma of local optimum. In order to show the effect of the optimization, we compare the support vector machine, the original random forest and the LGBM model, and perform Bayesian optimization on the latter two to achieve better results. Experiments show that the optimized model performs well in stock return prediction with a reduction of 0.01 in RMSE and MAE, a reduction of 0.4 in MAPE, and an improvement of 0.02 in R2 value compared to the other models. The Bayesian optimization based Random Forest model achieves the lowest prediction error on different datasets and has wide applicability.

关键词： Support vector machines Accuracy Time series analysis Human factors Predictive models Data models Bayes methods

来源：评论

学校读者我要写书评

暂无评论

Effect of the Mixing Structure Parameters of a Self-reflux Burner on Combustion Characteristics and NO_(x)Emission

引用

Journal of Thermal Science 2021年第4期30卷 1224-1236页

作者： XU Qian SHEN Meng SHI Kejian LIU Zhihui AKKURT Nevzat XIONG Yaxuan LIU Lin FENG Junxiao WANG Jiulong School of energy and environmental engineering Shunde Graduate SchoolUniversity of Science and Technology BeijingBeijing 100083China Beijing Key Laboratory of Energy Conservation and Emission Reduction for Metallurgical Industry Beijing 100083China Department of Mechanical Engineering Munzur UniversityTunceli 62000Turkey Key Laboratory of HVAC Beijing University of Civil Engineering and ArchitectureBeijing 100044China School of Mathematics and Physics University of Science and Technology BeijingBeijing 100083China Department of Big Data Technology and Application Development Computer Network Information CenterChinese Academy of SciencesBeijing 100190China

To solve the problem of low efficiency of NO_(x)emission reduction in self-reflux burners,this study is based on the concept of coordinated control of self-reflux burner structural and thermal *** completing the structural design and optimization of thermal parameters,we continue to adjust the two key structural parameters:the nozzle axis distance and the length of the cylindrical section,to minimize NO_(x)*** are the two parameters that chiefly affect the mixing of flue gas and fuel *** results show that increasing nozzle axis distance can delay the mixing of gas and air and create a more uniform oxygen concentration field for the combustion *** maximum combustion temperature is reduced from 1973.65 K to 1935.88 K and the volume fraction of NO_(x)in the flue gas is reduced from 188.08×10^(–6)to 143.47×10^(–6).However,compared with the nozzle axis distance,the length of the cylindrical section of the burner has little effect on the mixing of the flow *** different cylindrical section lengths,the maximum combustion temperature does not change more than 3 K,and the volume fraction of NO_(x)in the flue gas changes within 5×10~(–6).

关键词： self-reflux structure structural parameter control thermal parameter combustion characteristic NO_(x)emission

来源：评论

学校读者我要写书评

暂无评论

Trojan Horse Training for Breaking Defenses against Backdoor Attacks in Deep Learning

arXiv

引用

arXiv 2022年

作者： Rajabi, Arezoo Ramasubramanian, Bhaskar Poovendran, Radha Network Security Lab Department of Electrical and Computer Engineering University of Washington SeattleWA98195 United States Electrical and Computer Engineering Western Washington University BellinghamWA98225 United States

Machine learning (ML) models that use deep neural networks are vulnerable to backdoor attacks. Such attacks involve the insertion of a (hidden) trigger by an adversary. As a consequence, any input that contains the trigger will cause the neural network to misclassify the input to a (single) target class, while classifying other inputs without a trigger correctly. ML models that contain a backdoor are called Trojaned models. Backdoors can have severe consequences in safety-critical cyber and cyber physical systems when only the outputs of the model are available. Defense mechanisms have been developed and illustrated to be able to distinguish between outputs from a Trojaned model and a non-Trojaned model in the case of a single-target backdoor attack with accuracy > 96%. Understanding the limitations of a defense mechanism requires the construction of examples where the mechanism fails. Current single-target backdoor attacks require one trigger per target class. We introduce a new, more general attack that will enable a single trigger to result in misclassification to more than one target class. Such a misclassification will depend on the true (actual) class that the input belongs to. We term this category of attacks multi-target backdoor attacks. We demonstrate that a Trojaned model with either a single-target or multi- target trigger can be trained so that the accuracy of a defense mechanism that seeks to distinguish between outputs coming from a Trojaned and a non-Trojaned model will be reduced. Our approach uses the non-Trojaned model as a 'teacher' for the Trojaned model and solves a min-max optimization problem between the Trojaned model and defense mechanism. Empirical evaluations demonstrate that our training procedure reduces the accuracy of a state-of-the-art defense mechanism from > 96% to 0%. We also discuss possible approaches to improve defense mechanisms to ensure resilience to backdoor attacks for a broader category of ML models. © 2022, CC BY.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Multi-modal Data based Semi-Supervised Learning for Vehicle Positioning

arXiv

引用

arXiv 2024年

作者： Huan, Ouwen Yang, Yang Luo, Tao Chen, Mingzhe Beijing Laboratory of Advanced Information Network Beijing University of Posts and Telecommunications Beijing100876 China Department of Electrical and Computer Engineering Institute for Data Science and Computing University of Miami Coral GablesFL33146 United States Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China

In this paper, a multi-modal data based semisupervised learning (SSL) framework that jointly use channel state information (CSI) data and RGB images for vehicle positioning is designed. In particular, an outdoor positioning system where the vehicle locations are determined by a base station (BS) is considered. The BS equipped with several cameras can collect a large amount of unlabeled CSI data and a small number of labeled CSI data of vehicles, and the images taken by cameras. Although the collected images contain partial information of vehicles (i.e. azimuth angles of vehicles), the relationship between the unlabeled CSI data and its azimuth angle, and the distances between the BS and the vehicles captured by images are both unknown. Therefore, the images cannot be directly used as the labels of unlabeled CSI data to train a positioning model. To exploit unlabeled CSI data and images, a SSL framework that consists of a pretraining stage and a downstream training stage is proposed. In the pretraining stage, the azimuth angles obtained from the images are considered as the labels of unlabeled CSI data to pretrain the positioning model. In the downstream training stage, a small sized labeled dataset in which the accurate vehicle positions are considered as labels is used to retrain the model. Simulation results show that the proposed method can reduce the positioning error by up to 30% compared to a baseline where the model is not pretrained. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

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：