检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xinyu Zhang Qize Jiang Hanyuan Zhang Weiwei Sun School of Computer Science Shanghai Key Laboratory of Data Science Fudan University Shanghai China Shanghai Institute of Intelligent Electronics & Systems Shanghai China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Hypergraph neural networks can model more flexible connectivity relationships, are used to model higher-order interactions, and have produced strong results in many real-world applications. However, the currently existing hypergraph neural networks need more exploration in capturing the global positional information of nodes in hypergraphs. Although there have been many explorations of the problem in graph neural networks, extending these approaches to hypergraphs is fraught with challenges. The major challenge is that hyperedges in hypergraphs are the other dimensional element of the incidence structure, have more flexible definitions than edges in graphs, and require more attention when learning global positional information. We propose a novel position-aware hypergraph message-passing neural network framework to address the above challenges. Specifically, we propose a global positional embedding learning approach that can separately model global positional information for nodes and hyperedges. At the same time, we also optimize the learning of local structures with hyperedges. Experiments on several publicly available benchmark datasets find that our proposed method outperforms many state-of-the-art methods.

关键词： Neural networks Transforms Benchmark testing Signal processing Graph neural networks Acoustics Speech processing

来源：评论

学校读者我要写书评

暂无评论

UNeLF: Unconstrained Neural Light Field for Self-Supervised Angular Super-Resolution

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Zhao, Mingyuan Sheng, Hao Chen, Rongshan Cong, Ruixuan Cui, Zhenglong Yang, Da Beihang University State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beijing100191 China Beihang University Data Science and Intelligent Computing Laboratory Hangzhou International Innovation Institute Hangzhou311115 China Macao Polytechnic University Faculty of Applied Sciences 999078 China

Compared to supervised learning methods, self-supervised learning methods address the domain gap problem between light field (LF) datasets collected under varying acquisition conditions, which typically leads to decreased performance when differences exist in the distribution between the training and test sets. However, current self-supervised light field angular super-resolution (LFASR) techniques primarily focus on exploiting discrete spatial-angular features while neglecting continuous LF information. In contrast to previous work, we propose a self-supervised unconstrained neural light field (UNeLF) to continuously represent LF for LFASR. Specifically, any LF can be described as the camera pose for each sub-aperture image (SAI) and the two-plane that captures these SAIs. To describe the former, we introduce a SAIs-dependent pose optimization method to solve the issue that arises from the narrow baseline of most LF data, which hinders robust camera pose estimation. This mechanism reduces the number of trainable camera parameters from a quadratic to a constant scale, thereby alleviating the complexity of joint optimization. For the latter, we propose a novel adaptive two-plane parameterization strategy to determine the two-plane that captures these SAIs, facilitating refocusing. Finally, we jointly optimize the camera parameters, near-far planes and neural light field, efficiently mapping each adaptive two-plane parameterized ray to its correspondence color in a continuous manner. Comprehensive experiments demonstrate that UNeLF achieves faster training and inference with fewer computational resources while exhibiting superior performance on both synthetic and real-world datasets. © 1991-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Spatial-Temporal Resource Utilization for Partially Connected Multi-AUV Broadcast Scheduling

引用

IEEE Wireless Communications Letters 2025年第5期14卷 1321-1325页

作者： He, Hongyi Du, Jun Jiang, Chunxiao Wang, Jintao Song, Jian Niyato, Dusit Tsinghua University Beijing National Research Center for Information Science and Technology State Key Laboratory of Space Network and Communications Department of Electronic Engineering Beijing100084 China Tsinghua University Department of Electronic Engineering State Key Laboratory of Space Network and Communications Beijing100084 China Tsinghua University Beijing National Research Center for Information Science and Technology Beijing100084 China Tsinghua University Beijing National Research Center for Information Science and Technology Department of Electronic Engineering Beijing100084 China Tsinghua University Shenzhen International Graduate School Shenzhen518055 China Nanyang Technological University College of Computing and Data Science Jurong West Singapore

The long propagation delays in underwater acoustic channels can be leveraged to improve communication efficiency using spatial-temporal reuse techniques. This letter focuses on a system of partially connected AUVs that continuously broadcast packets. A central controller gathers the location information of the AUVs and designs a broadcast sequence exploiting propagation delays to minimize the broadcast cycle. Since finding optimal solutions is time-consuming, we propose a scheduling algorithm that uses heuristics based on partial connections to reduce computational complexity. Additionally, we allow collision tolerance to shorten the broadcast cycle further and investigate the trade-off of the shorter broadcast cycle and collision probability, resulting in a scheduling algorithm exhibiting receiving-alignment properties. Simulations demonstrate the effectiveness of the proposed algorithm in reducing broadcast cycles while maintaining low computational complexity. © 2012 IEEE.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Hybrid genetic algorithm with rank sorting for obtaining bijective S-boxes with good cryptographic properties 4

Hybrid genetic algorithm with rank sorting for obtaining bij...

引用

4th International Conference on Network Communication and Information Security, ICNCIS 2024

作者： Li, Xin Liu, Zhenghong Zhao, Haixia School of Mathematics and Computing Science Guilin University of Electronic Technology Guilin541004 China Guilin541004 China Guangxi Colleges Universities Key Laboratory of Data Analysis and Computation Guilin541004 China School of Information and Communication Guilin University of Electronic Technology Guilin541004 China

ISBN: (数字)9781510688261

ISBN: (纸本)9781510688254

An S-box is an essential component in block ciphers, and its cryptography properties play a significant role in the security of the cipher. A good S-box must have high nonlinearity (NLF ) and low differential uniformity (χF ). In this paper, a rank-sorting hybrid genetic algorithm is designed to find S-boxes that possess high nonlinearity and low difference uniformity. Compared with the traditional genetic algorithm, the rank-sorting hybrid genetic algorithm designed in this work was improved in the following aspects. First, the crossover protection strategy was designed and used in the crossover phase, which can assure the bijectivity of offspring. Second, an elite selection strategy based on the value of the fitness function was utilized in the selection step to prevent the loss of excellent individuals. In particular, a sorting method named the rank-sorting algorithm was proposed and implemented on the selected offspring before the next iteration began. The result of using the rank-sorting algorithm is that the differences between individuals for the next crossover increase, the possibility of generating excellent offspring increases, and the risk of the whole algorithm becoming stuck in local optimal solutions decreases. The simulation results show that when the number of variables is even (8 to 12), the bijective S-boxes with almost high nonlinearity and low differential uniformity can be searched by using the rank-sorting hybrid genetic algorithm. © 2025 SPIE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Large-Scale Riemannian Meta-Optimization via Subspace Adaptation

arXiv

引用

arXiv 2025年

作者： Yu, Peilin Wu, Yuwei Gao, Zhi Fan, Xiaomeng Jia, Yunde Beijing Key Laboratory of Intelligent Information Technology School of Computer Science & Technology Beijing Institute of Technology China Guangdong Laboratory of Machine Perception and Intelligent Computing Shenzhen MSU-BIT University China

Riemannian meta-optimization provides a promising approach to solving non-linear constrained optimization problems, which trains neural networks as optimizers to perform optimization on Riemannian manifolds. However, existing Riemannian meta-optimization methods take up huge memory footprints in large-scale optimization settings, as the learned optimizer can only adapt gradients of a fixed size and thus cannot be shared across different Riemannian parameters. In this paper, we propose an efficient Riemannian meta-optimization method that significantly reduces the memory burden for large-scale optimization via a subspace adaptation scheme. Our method trains neural networks to individually adapt the row and column subspaces of Riemannian gradients, instead of directly adapting the full gradient matrices in existing Riemannian meta-optimization methods. In this case, our learned optimizer can be shared across Riemannian parameters with different sizes. Our method reduces the model memory consumption by six orders of magnitude when optimizing an orthogonal mainstream deep neural network (e.g. ResNet50). Experiments on multiple Riemannian tasks show that our method can not only reduce the memory consumption but also improve the performance of Riemannian meta-optimization. © 2025, CC BY-NC-ND.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

ICPR 2024 Competition on Beyond Visible Spectrum: AI for Agriculture 27th

ICPR 2024 Competition on Beyond Visible Spectrum: AI for A...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Han, Liangxiu Huang, Wenjiang Zhang, Xin Dong, Yingying Sobeih, Tam Lin, Yufan Department of Computing and Mathematics Manchester Metropolitan University ManchesterM15 6BH United Kingdom Key laboratory of Digital Earth Science Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China Institute of Data Science National Cheng Kung University Taiwan70101 China

ISBN: (纸本)9783031801389

The ICPR 2024 Competition on "Beyond Visible Spectrum: AI for Agriculture" presents an exciting opportunity for researchers and practitioners to advance computer vision techniques in agricultural crop disease monitoring. By leveraging the vast multi/hyper spectral remote sensing datasets, participants are encouraged to develop innovative machines/deep learning algorithms. The primary objective is to improve the accuracy and efficiency of crop disease identification, thereby fostering sustainable agricultural practices and facilitating global knowledge generation for enhanced food security. This challenge presents participants with the opportunity to make significant contributions to the advancement of precision farming and crop monitoring techniques. By improving the machines/deep learning models, the competition aims to drive forward the capabilities of agricultural technology, making it more efficient, accurate, and sustainable. A total of 190 people from more than 10 countries in Europe, Asia and North America entered the competition by the deadline date, with a total of 126 submissions of results. The rank1 team obtained an accuracy of 73.3% in crop disease classification task with hyperspectral dataset. In the self-supervised learning task, the rank1 team obtained 79.2% accuracy. We report the leader-boards of the competition in this report and present three innovative approaches of the participants. The ICPR 2024 Competition on "Beyond Visible Spectrum" is not just a competition;it is a call to action for the computer research community to bring about transformative changes in how agricultural data is analyze and utilized, paving the way for smarter, more sustainable farming practices that can benefit communities worldwide. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Semantic Pre-Extraction for Energy-Efficient AoI Minimization in UAV-Assisted Wireless Networks

Semantic Pre-Extraction for Energy-Efficient AoI Minimizatio...

引用

IEEE Conference on Wireless Communications and Networking

作者： Yusi Long Gary C.F. Lee Lanhua Li Shimin Gong Sumei Sun Dusit Niyato School of Intelligent Systems Engineering Shenzhen Campus of Sun Yat-sen University China Guangdong Provincial Key Laboratory of Fire Science and Intelligent Emergency Technology China Institute for Infocomm Research (IR) A*STAR Singapore College of Computing and Data Science Nanyang Technological University Singapore

ISBN: (数字)9798350368369

ISBN: (纸本)9798350368376

This paper investigates an unmanned aerial vehicle (UAV)-assisted semantic communication network. The energy-limited ground users (GUs) provide semantic services to periodically generated raw data and a UAV relays the extracted semantic information to a base station (BS). Semantic extraction enhances data responsiveness and reduces the age-of-information (AoI) by transmitting only the most essential information. However, more complex semantic extraction increases energy consumption, making it easier for the GUs to deplete their energy. Therefore, we introduce a novel energy-efficient AoI (EAoI) metric to capture both information freshness and energy consumption of the GUs. We formulate a time-averaged EAoI minimization problem by jointly optimizing the GUs' scheduling, pre-extraction strategy, semantic control, computing resource allocation, and the UAV's trajectory. We further propose a semantic-aware joint pre-extraction and trajectory planning (Sem-JPT) algorithm to decompose the complex optimization problem into three subproblems, which are solved by a series of approximation methods. Simulation results demonstrate that semantic communication can reduce the overall EAoI by more than 18% compared with conventional bit-based communication. Moreover, the proposed Sem-JPT algorithm can maintain information freshness and prolong the GUs' lifetimes, outperforming existing baselines.

关键词： Energy consumption Trajectory planning Processor scheduling Wireless networks Semantic communication Minimization Information age Energy efficiency Trajectory data mining

来源：评论

学校读者我要写书评

暂无评论

data-Driven Agent-Based Model for Public Opinion Propagation Simulation in Cyberbullying

引用

Big data Mining and Analytics 2025年第4期8卷 794-819页

作者： Wang, Runyu Lu, Tun Zhang, Peng Gu, Ning School of Computer Science Fudan University Shanghai 200433 China School of Computer Science Fudan Institute on Aging Fudan University Shanghai Key Laboratory of Data Science MOE Laboratory for National Development and Intelligent Governance Shanghai 200433 China Shanghai Institute of Intelligent Electronics & Systems Fudan University Shanghai 200433 China

The flourishing development of social media platforms based on cultivating user relationships to spread and share information has provided a breeding ground for cyberbullying. How to infer the evolution of public opinion propagation in an online bullying environment is of great significance to the governance of cyberbullying. In this paper, we propose a data-driven agent-based Model for Public Opinion Propagation Simulation (MPOPS) in cyberbullying. First, we design a Public Opinion Propagation Environment for Opinion Fusion and Polarization (OFP-POPE) in a cyber-physical-social space. Second, we conduct agent-based fine-grained modeling based on the OFP-POPE. Third, we define and quantify user interaction behaviors and improve the Susceptible-Exposed-Infected Removed (SEIR) model by taking these interaction behaviors as the factors influencing public opinion propagation. Finally, we take the “2022 Tangshan restaurant attack” incident as an empirical study case and conduct simulation experiments on the MPOPS driven by real-world data. The experimental results show that the MPOPS is superior to other baseline models, and can simulate the evolutionary trend of public opinion propagation in actual cyberbullying scenarios. In addition, the ablation experiment and parameter sensitivity analysis provide a reference for future cyberbullying intervention efforts. © 2018 Tsinghua University Press.

关键词： agent-based modeling cyberbullying opinion dynamics public opinion social media platform

来源：评论

学校读者我要写书评

暂无评论

Riemannian Complex Hermit Positive Definite Convolution Network for Polarimetric SAR Image Classification

arXiv

引用

arXiv 2025年

作者： Shi, Junfei Nie, Mengmeng Li, Yuke Jin, Haiyan Lin, Weisi Department of Computer Science and Technology Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an University of Technology Xi’an China College of Computing and Data Science Nanyang Technological University 639798 Singapore

Deep learning can learn high-level semantic features in Euclidean space effectively for PolSAR images, while they need to covert the complex covariance matrix into a feature vector or complex-valued vector as the network input. However, the complex covariance matrices are essentially a complex Hermit positive definite (HPD) matrix endowed in Riemannian manifold rather than Euclidean space. The matrix’s real and imagery parts are with the same significance, as the imagery part represents the phase information. The matrix vectorization will destroy the geometric structure and manifold characteristics of complex covariance matrices. To learn complex HPD matrices directly, we propose a Riemannian complex HPD convolution network(HPD CNN) for PolSAR images. This method consists of a complex HPD unfolding network(HPDnet) and a CV-3DCNN enhanced network. The proposed complex HPDnet defines the HPD mapping, rectifying and the logEig layers to learn geometric features of complex matrices. In addition, a fast eigenvalue decomposition method is designed to reduce computation burden. Finally, a Riemannian-to-Euclidean enhanced network is defined to enhance contextual information for classification. Experimental results on two real PolSSAR datasets demonstrate the proposed method can achieve superior performance than the state-of-the-art methods especially in heterogeneous regions. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Transductive Few-shot Learning via Joint Message Passing and Prototype-based Soft-label Propagation

引用

ACM Transactions on Multimedia computing, Communications and Applications 2025年第4期21卷

作者： Wang, Jiahui Xu, Qin Jiang, Bo Luo, Bin The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China The Information Materials and Intelligent Sensing Laboratory of Anhui Province School of Computer Science and Technology Anhui University The Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei China

The transductive Few-shot Learning (FSL) mostly employs either prototype learning or label propagation methods to generalize to new classes by using the information of all query samples. However, existing methods have several main limitations. First, the prototype methods mainly focus on support samples which fail to fully exploit the relationships of query samples. Second, existing label propagation methods are generally not effective for the class-imbalanced problem. Third, existing works usually optimize the learnable parameters during inference which significantly reduces the efficiency of existing methods. To address these limitations, this article proposes an efficient and robust method for transductive FSL problem, termed Prototype-based Soft-label Propagation (PSLP), which combines the prototype learning and label propagation together for FSL problem. In our proposed method, first, the soft-label presentation for each query sample is estimated by leveraging prototypes. Then, the soft-label propagation is conducted on the learned query-support graph and the prototype representation is rectified. Both steps are conducted progressively for boosting the performance. Moreover, to learn effective prototypes for soft-label estimation and the desirable query-support graph for soft-label propagation, we design a new joint message passing scheme to learn the sample presentation and relational graph jointly. The PSLP method is parameter-free and can be implemented very efficiently. The experiments conducted on four popular benchmarks show that our method achieves competitive results on both balanced and imbalanced settings compared to the state-of-the-art methods. The code is released at https://***/mobulan/PSLP. © 2025 Copyright held by the owner/author(s).

关键词： Few-shot learning Transductive learning Message passing Label propagation

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

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

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：