检索结果-内蒙古大学图书馆

IEEE International Symposium on information Theory

作者： Yi Song Kai Wan Zhenyu Liao Hao Xu Giuseppe Caire Shlomo Shamai Faculty of Electrical Engineering and Computer Science Technical University of Berlin Berlin Germany School of Electronic Information and Communications Huazhong University of Science and Technology Wuhan China Department of Electronic and Electrical Engineering University College London London UK Department of Electrical and Computer Engineering Technion-Israel Institute of Technology Haifa Israel

ISBN: (数字)9798350382846

ISBN: (纸本)9798350382853

In this paper, we consider a remote source coding problem with binary phase shift keying (BPSK) modulation sources, where observations are corrupted by additive white Gaussian noise (AWGN). An intermediate node, such as a relay, receives these observations and performs further compression to find the optimal trade-off between complexity and relevance. This problem can be formulated as an information bottleneck (IB) problem with Bernoulli sources and Gaussian mixture observations, for which no closed-form solution is known. To address this challenge, we propose a unified achievable scheme that employs three different compression strategies for intermediate node processing, i.e., two-level quantization, multi-level deterministic quantization, and soft quantization with tanh function. Comparative analyses with existing methods, such as the Blahut-Arimoto (BA) algorithm and the information Dropout approach, are performed through numerical evaluations. The proposed analytic scheme is observed to consistently approach the (numerically) optimal performance over a range of signal-to-noise ratios (SNRs), confirming its effectiveness in the considered setting.

关键词： Quantization (signal) Closed-form solutions AWGN Phase modulation Source coding Approximation algorithms Binary phase shift keying

来源：评论

学校读者我要写书评

暂无评论

Large Capacity Image Watermarking Model using SWT and Laplacian Pyramid

Large Capacity Image Watermarking Model using SWT and Laplac...

引用

IEEE International Conference on High Performance Computing and Communications (HPCC)

作者： Baowei Wang Fan Yang Yufeng Wu Changyu Dai Wenjue Huang Xingyuan Zhao Engineering Research Center of Digital Forensics Ministry of Education China Waterford Institute Nanjing University of Information Science and Technology Nanjing China School of Computer and Software Nanjing University of Information Science and Technology Nanjing China

Image watermarking algorithms have developed rapidly in recent years. Most of the Deep Neural Network(DNN)-based image watermarking algorithms embed only 30 bits or 64 bits messages in 128 × 128 images, while few algorithms embed large capacity messages. This is mainly because embedding large capacity messages will significantly influence the performance of algorithms. However, the limited capacity of these messages inhibits their effectiveness in certain data protection scenarios. Therefore, we propose a DNN-based large capacity image water-marking model. By using Stationary Wavelet Transform(SWT) and Laplacian pyramid transform, we expand the embedding capacity and achieve embedding 128 bits messages in 128 × 128 images. Experimental results show that under the same conditions, the proposed model is superior to state-of-the-art models, and has excellent performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Semi-Supervised Deep Nonnegative Matrix Factorization with Constraint Propagation for Data Representation

引用

IEEE Transactions on Big Data 2025年

作者： Peng, Siyuan Yin, Jingxing Yang, Zhijing Nie, Feiping Chen, Badong Guangdong University of Technology School of Information Engineering 510006 China Northwestern Polytechnical University School of Computer Science Center for OPTical IMagery Analysis and Learning Xi'an710072 China Xi'an Jiaotong University Institute of Artificial Intelligence and Robotics Xi'an710049 China

Deep nonnegative matrix factorization (DNMF) technique has attached much great attention in recent year, since it can effectively discover the underlying hierarchical structure of complex data. However, most existing unsupervised and semi-supervised DNMF approaches not only suffer from the noisy data seriously, but also fail to enhance the decomposition quality of DNMF obviously by using the obtained supervisory information. To overcome these drawbacks, a robust semi-supervised DNMF method, called the correntropy based semi-supervised DNMF with constraint propagation (CSDCP), is proposed in this paper for learning a compact and meaningful data representation from the original data. Particularly, instead of adopting the traditional Frobenius norm, CSDCP employs the nonlinear and local similarity measure (e.g., correntropy) as the loss function in DNMF to enhance the robustness of DNMF for the noisy data. In addition, the hypergraph based constraint propagation (HCP) algorithm is adopted in CSDCP to exploit the limited supervisory information fully for capturing good data representation. Moreover, algorithm analysis of CSDCP is presented in this paper, including convergence analysis, robustness analysis supervised information analysis, and computational complexity. Extensive experimental results have illustrated that, in comparison to the most related DNMF approaches, CSDCP usually has better clustering results on six nonnegative datasets in clustering tasks. © 2015 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Physics-Informed Neural Networks with Complementary Soft and Hard Constraints for Solving Complex Boundary Navier-Stokes Equations

arXiv

引用

arXiv 2024年

作者： Zhou, Chuyu Li, Tianyu Lan, Chenxi Du, Rongyu Xin, Guoguo Nan, Pengyu Yang, Hangzhou Wang, Guoqing Liu, Xun Li, Wei School of Physics Northwest University Xi’an710127 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China Department of Imaging Technology Beijing Institute of Space Mechanics and Electricity Beijing100094 China School of Information Science & Technology Northwest University Xi’an710127 China

Soft- and hard-constrained Physics Informed Neural Networks (PINNs) have achieved great success in solving partial differential equations (PDEs). However, these methods still face great challenges when solving the Navier-Stokes equations (NSEs) with complex boundary conditions. To address these challenges, this paper introduces a novel complementary scheme combining soft and hard constraint PINN methods. The soft-constrained part is thus formulated to obtain the preliminary results with a lighter training burden, upon which refined results are then achieved using a more sophisticated hard-constrained mechanism with a primary network and a distance metric network. Specifically, the soft-constrained part focuses on boundary points, while the primary network emphasizes inner domain points, primarily through PDE loss. Additionally, the novel distance metric network is proposed to predict the power function of the distance from a point to the boundaries, which serves as the weighting factor for the first two components. This approach ensures accurate predictions for both boundary and inner domain areas. The effectiveness of the proposed method on the NSEs problem with complex boundary conditions is demonstrated by solving a 2D cylinder wake problem and a 2D blocked cavity flow with a segmented inlet problem, achieving significantly higher accuracy compared to traditional soft- and hard-constrained PINN approaches. Given PINN’s inherent advantages in solving the inverse and the large-scale problems, which are challenging for traditional computational fluid dynamics (CFD) methods, this approach holds promise for the inverse design of required flow fields by specifically-designed boundary conditions and the reconstruction of large-scale flow fields by adding a limited number of training input points. Copyright © 2024, The Authors. All rights reserved.

关键词： Computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

Beyond the Label Itself: Latent Labels Enhance Semi-supervised Point Cloud Panoptic Segmentation

arXiv

引用

arXiv 2023年

作者： Chen, Yujun Tan, Xin Zhang, Zhizhong Qu, Yanyun Xie, Yuan School of Computer Science and Technology East China Normal University China Chongqing Institute East China Normal University China School of Information Science and Engineering Xiamen University China

As the exorbitant expense of labeling autopilot datasets and the growing trend of utilizing unlabeled data, semi-supervised segmentation on point clouds becomes increasingly imperative. Intuitively, finding out more "unspoken words" (i.e., latent instance information) beyond the label itself should be helpful to improve performance. In this paper, we discover two types of latent labels behind the displayed label embedded in LiDAR and image data. First, in the LiDAR Branch, we propose a novel augmentation, Cylinder-Mix, which is able to augment more yet reliable samples for training. Second, in the Image Branch, we propose the Instance Position-scale Learning (IPSL) Module to learn and fuse the information of instance position and scale, which is from a 2D pre-trained detector and a type of latent label obtained from 3D to 2D projection. Finally, the two latent labels are embedded into the multi-modal panoptic segmentation network. The ablation of the IPSL module demonstrates its robust adaptability, and the experiments evaluated on SemanticKITTI and nuScenes demonstrate that our model outperforms the state-of-the-art method, LaserMix. Copyright © 2023, The Authors. All rights reserved.

关键词： Optical radar

来源：评论

学校读者我要写书评

暂无评论

Efficient Congestion Control Through Energy and Traffic-Aware Clustering In Future Wireless Sensor Networks

Efficient Congestion Control Through Energy and Traffic-Awar...

引用

International Conference on Futuristic Technologies (INCOFT)

作者： Vennila N L Saju Raj T Aparna Atul Junnarkar K.B.Devi Shalini Geetha P Department of Computer Science Muslim Arts College Thiruvithancode Kanyakumari India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Avadi Chennai India Department of Information Technology Vishwakarma Institute Of Information Technology. (VIIT) Pune India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences (SIMATS) Thandalam Chennai India

When the aggregated capacity of a channel is surpassed by the amount of data flowing across it, a condition known as congestion occurs in wireless sensor networks (WSNs). Because retransmission of every unacknowledged packet is not an optimal solution in terms of energy for resource-restricted sensor nodes, the result is overflowing of the buffer at each receiving sensor node. This eventually leads to a reduction in the packet delivery ratio, the dropping of the packets, and a degradation of the network throughput. Since the routing problem has been shown to be NP-hard and it has been revealed that a heuristic-based method outperforms their traditional counterparts, routing is one of the most common strategies for lowering node energy consumption and boosting throughput in WSNs. In addition, routing is one of the most popular ways for raising throughput. As a result, there is a need for optimum routing solutions for the WSNs of the future, taking into consideration both the energy and traffic factors. This research will only focus on wireless sensor networks (WSNs for short). The suggested protocol will be referred to as Energy and Traffic-aware Clustering for Congestion Control (ETCCC) throughout this article. In graphs compared ETCCC performances with three recent protocol names.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Leveraging Machine Learning for Dynamic Resource Allocation in Cloud Automation

Leveraging Machine Learning for Dynamic Resource Allocation ...

引用

Automation and Computation (AUTOCOM), International Conference on

作者： Gunveen Ahluwalia Vaibhav Goutham Suresh M D Anto Praveena Kalyan Acharjya Trushna Paridaname Centre of Research Impact and Outcome Chitkara University Rajpura Punjab India School of Mangement - PG Management JAIN (Deemed to be University) Bangalore Karnataka Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai Tamil Nadu India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India Department of Computer Science and Engineering Siksha ‘O’ Anusandhan (Deemed to be University) Bhubaneswar Odisha India

ISBN: (数字)9798331542375

ISBN: (纸本)9798331542382

Cloud automation enables the dynamic management and allocation of resources in a computing environment and is one of the most common uses available. Yet despite their power when cloud systems were simple and small, traditional approaches to allocating resources have proven far less feasible lately. This scenario has also created the need for new evolution methods that will adapt to dynamic cloud environments and improve the system's overall performance. A solution to this challenge comes from machine learning, which has been increasingly used recently. ML algorithms can process vast amounts of data and identify patterns and trends to predict resource use and user needs. Deep learning helps cloud-enabled automation systems adapt in real-time to changing conditions, distribute resources efficiently and improve user experience. For example, reinforcement learning and deep learning are ML techniques that can be applied to resource allocation algorithms or demand prediction models to increase accuracy. Integrating cloud automation and machine learning brings effectiveness in dynamic resource allocation, boosting the overall performance of a system, which leads to savings, better experience, and greater flexibility in using cloud resources. Reinforcement and deep learning are machine learning techniques to optimize order allocation algorithms and enhance demand estimation models.

关键词： Deep learning Cloud computing Automation Machine learning algorithms Heuristic algorithms Estimation Dynamic scheduling Prediction algorithms Real-time systems Resource management

来源：评论

学校读者我要写书评

暂无评论

Interpreting the Evidence on Life Cycle to Improve Educational Outcomes of Students Based on Generalized ARC-GRU Approach 3

Interpreting the Evidence on Life Cycle to Improve Education...

引用

3rd International Conference for Innovation in technology, INOCON 2024

作者： Madumidha, S. Tripathi, Anju Srinivasan, P. Wategaonkar, Somnath Raghunath Chauhan, Amit Suganthi, D. Sri Krishna College of Technology Department of Information Technology Coimbatore India Mangalmay Institute of Management and Technology Dehi NCR Gr.Noida India Sona College of Technology Department of Ece Tamilnadu Salem India Bharati Vidyapeeth College of Engineering Department of Electronics and Telecommunication Navi Mumbai India School of Sciences Department of Life Sciences Karnataka Bengaluru India Saveetha College of Liberal Arts and Sciences Simats Department of Computer Science Chennai India

ISBN: (纸本)9798350381931

Research on the effects of teachers' fatigue on students' learning has been significantly less common than research on the effects of teachers' fatigue on teachers' own performance. Therefore, the purpose of this research is to see if teachers' emotional weariness has any bearing on their students' performance in the classroom. Consideration is given to a student's grades and their impressions of whether or not the system receive assistance from teachers, as well as to the student's general outlook on school, confidence in their own abilities, and faith in the availability of faculty support. Data preparation, feature extraction, and model training are the first steps in the proposed approach. Indicators of the quality of the education being provided are eliminated (by outlier removal and feature scaling). k-mean clustering approach is a technique of clustering which is commonly used in feature extraction. Following feature extraction, GARCH-GRU models are trained. The proposed approach is superior to two popular alternatives, ARCH and GRU. Using the provided method, the system were able to achieve a maximum accuracy of 97.07%. © 2024 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Enhancing Hypertrophic Cardiomyopathy Heart Risk Assessment with Deep Learning and Multi-Omics Data Integration

Enhancing Hypertrophic Cardiomyopathy Heart Risk Assessment ...

引用

Artificial Intelligence and Emerging technology (Global AI Summit), International Conference on

作者： Afreen Fatima Mohammed Dinesh Prasad Sahu Rinki Tyagi Prachi Pundhir Rami Riad Hussien Surbhi Kamboj Department of CSE-DS CVR College of Engineering Hyderabad India School of Computer Science Engineering & Technology Bennett University Greater Noida India Department of Computer Science and Engineering Raj Kumar Goel Institute of Technology Ghaziabad India Department of Information Technology ABES Engineering College Ghaziabad India The Islamic University Najaf Iraq KIET School of Pharmacy KIET Group of Institutions Ghaziabad India

ISBN: (数字)9798350379716

ISBN: (纸本)9798350379723

This research explores the possibility of improving risk assessment for hypertrophic cardiomyopathy (HCM) by combining multi-omics data and deep learning methodologies. In this analysis, the research included 513 patients and established prediction models based on genomics, transcriptomics, and proteomics data. The research also used several deep learning architectures, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory networks (LSTM), and graph neural network solvers. The notable results were that the CNN model's accuracy was 93%, with corresponding precision and recall rates of 91% and 92%, respectively. In addition, the performance of the models was increased when combining different omics data hierarchically. CNNs showed improvements in accuracy from 85% using single-omics to 92% with integrated multi-omics. Revealing the intricate biological relationships associated with Hypertrophic cardiomyopathy (HCM) necessitates high-quality, comprehensive data integration. These insights not only contribute to a more comprehensive understanding of HCM's pathophysiology but also serve as seeds for the discovery of novel biomarkers and new therapeutic targets. In summary, this work demonstrates the power of multi-omics and deep learning approaches for developing precision medicine in HCM with applications to improved risk stratification and patient management strategies.

关键词： Deep learning Accuracy Recurrent neural networks Transcriptomics Proteomics Biomarkers Data models Risk management Convolutional neural networks Long short term memory

来源：评论

学校读者我要写书评

暂无评论

A Spectral Learning Enhancement method in Cooperative multiagent system for production control system

A Spectral Learning Enhancement method in Cooperative multia...

引用

Intelligent Algorithms for Computational Intelligence Systems (IACIS), International Conference on

作者： V. Sangeetha S Padmakala Sutha Merlin J S. Prabu K G Parthiban Dept of CSE Saveetha School of Engineering Chennai India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences SIMATS Chennai India Department of Information Technology S.A.Engineering College Chennai India Department of ECE Mahendra Institute of Technology Namakkal India Department of Biomedical Engineering Dhaanish Ahmed Institute of Technology Coimbatore India

ISBN: (数字)9798350360660

ISBN: (纸本)9798350360677

In modern production control systems, efficient management of resources and coordination among multiple agents are crucial for optimizing performance and achieving operational goals. This study introduces the Spectral Learning Enhancement (SLE) method, a novel approach designed to improve decision-making and coordination in cooperative multi-agent systems. The SLE method leverages spectral learning techniques to analyze and enhance the communication and interaction patterns between agents, leading to more informed and adaptive production control strategies. The SLE method integrates spectral graph theory with advanced learning algorithms to model the relationships between agents and their environment. By analyzing the spectral properties of the interaction graph, the method identifies key patterns and dependencies that inform the learning process. This approach allows for the dynamic adjustment of agent behaviors and strategies based on real-time feedback and evolving production conditions. Results from extensive simulations and real-world case studies demonstrate the efficacy of the SLE method. In a controlled production environment, the application of SLE led to a 15% increase in overall system efficiency, as measured by production output and resource utilization. Additionally, the method improved coordination among agents, reducing communication overhead by 20% and minimizing production downtime by 10%. The enhanced adaptability of the agents, driven by the spectral learning framework, resulted in a 25% reduction in response time to unexpected changes in production demands. The SLE method's ability to enhance both individual and collective agent performance positions it as a significant advancement in the field of production control. By incorporating spectral analysis into multi-agent systems, the method provides a robust and scalable solution for optimizing production processes and achieving greater operational efficiency.

关键词： Measurement Accuracy Decision making Production control Graph theory Time factors Synchronization Resource management Spectral analysis Multi-agent 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：